Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

UK clinical practice guidelines for the management of gastrointestinal stromal tumours (GIST)

UK clinical practice guidelines for the management of gastrointestinal stromal tumours (GIST) Background: Soft tissue sarcomas (STS) are rare tumours arising in mesenchymal tissues. Gastrointestinal stromal tumour (GIST ) is the commonest STS and arises within the wall of the gastrointestinal (GI) tract. While most GISTs occur in the stomach they do occur in all parts of the GI tract. As with other STS, it is important that GISTs are man- aged by expert teams, to ensure consistent and optimal treatment, as well as recruitment to clinical trials, and the ongoing accumulation of further knowledge of the disease. The development of appropriate guidance, by an expe- rienced panel referring to the evidence available, is therefore a useful foundation on which to build progress in the field. Methodology: British Sarcoma Group guidelines for the management of GIST were initially developed by a panel of physicians experienced in the management of GIST. This current version has been updated and amended with refer- ence to other European and US guidance. We have received input from representatives of all diagnostic and treat- ment disciplines as well as patient representatives. Levels of evidence and strength of recommendation gradings are those used by ESMO adapted from those published by the Infectious Disease Society of America. Conclusions: The guidelines cover aetiology, genetics and underlying molecular mechanisms, diagnosis and initial investigations, staging and risk stratification, surgery, neoadjuvant and adjuvant therapy, the management of advanced disease and follow-up. The importance of mutational analysis in guiding treatment is highlighted, since this can indicate the most effective treatment and avoid administration of ineffective drugs, emphasising the need for management in specialist centres. Background Incidence British Sarcoma Group guidelines for the management Gastrointestinal stromal tumours (GISTs) are rare of GIST were initially developed in 2004 by a panel of cancers, with an estimated unadjusted incidence of physicians experienced in the management of GIST rep- 1.5/100,000/year [3]. Data from the Rhȏne-Alpes region resenting all diagnostic and treatment disciplines. These of France [4] and “NHS England Cancer Registry” (per- Clinical Practice Guidelines update reflect the improve - sonal communication) suggest an incidence of just under ments in our knowledge of the disease and developments 11 per million per annum, equating to 650 clinically that have taken place since then [1]. Levels of evidence meaningful new cases a year in the UK, approximately and strength of recommendation gradings are those used 900 in total. Accurate data on prevalence in the UK are by ESMO adapted from those published by the Infectious not yet available. Disease Society of America (Table 1) [2]. The median age at diagnosis is around 60–65  years, with a wide range. Occurrence in children, adolescents and younger patients is very rare, although paediatric *Correspondence: Ian.Judson@icr.ac.uk GISTs represent a distinct subset, marked by female pre- The Institute of Cancer Research, Royal Marsden NHS Foundation Trust, dominance, absence of KIT/platelet-derived growth fac- Fulham Road, London SW3 6JJ, UK Full list of author information is available at the end of the article tor alpha (PDGFRA) mutations, usually gastric origin, © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Judson et al. Clin Sarcoma Res (2017) 7:6 Page 2 of 10 Table 1 Levels of evidence and Grades of recommendation leading to a dyad of GIST and paraganglioma [10, 11]. Levels of evidence • Type-1 neurofibromatosis, i.e. associated with loss of I Evidence from at least one large randomised, controlled trial of good function of NF1, whether sporadic or inherited, and methodological quality (low potential for a bias) or meta-analyses of well-conducted randomised trials without heterogeneity absence of mutations in KIT or PDGFRA, the GISTs are often multifocal, predominantly located in the II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or small bowel [12]. of trials with demonstrated heterogeneity • Familial GIST, i.e. families with a germ-line autoso- III Prospective cohort studies mal dominant mutation of KIT, are extremely rare, IV Retrospective cohort studies or case–control studies presenting with multiple GISTs at an early age. V Studies without control group, case reports, and experts’ opinions Grades of recommendation Diagnosis A Strong evidence for efficacy with a substantial clinical benefit, Clinical presentation and investigation strongly recommended The most common symptoms of GIST include upper gas - B Strong or moderate evidence for efficacy but with a limited clinical trointestinal bleeding and anaemia, whilst larger tumours benefit, generally recommended may present with abdominal pain/discomfort and a pal- C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs,…), optional pable mass. Small bowel GISTs may remain silent for a D Moderate evidence against efficacy or for adverse outcome, generally long period before presenting with an acute event such not recommended as haemorrhage or rupture. Symptomatic colorectal E Strong evidence against efficacy or for adverse outcome, never GISTs may present with abdominal pain, obstruction recommended and lower gastrointestinal bleeding; oesophageal and gastro-oesophageal junction GISTs with dysphagia. Some patients may have non-specific systemic symptoms such as weight loss, night sweats and fever. Lack of awareness often multifocal, and with possible lymph node metasta- of the presenting features may lead to delayed diagnosis ses [5]. of GIST in some patients. Small GISTs may be asympto- matic and are often diagnosed incidentally during inves- Aetiology tigation for other conditions. In most cases the aetiology is unknown, although it is Small lesions below 2 cm in diameter may safely be fol- reported that patients with GIST are more likely to be lowed by endoscopic ultrasound on an annual basis and diagnosed with another cancer than the general popu- biopsied or excised if they continue to grow. For larger lation [6, 7], suggesting a likely link with inherited lesions it is necessary to make a histological diagnosis. The increased susceptibility to cancer in some patients. In standard approach for small nodules of 2  cm or above is the majority of cases GIST is associated with an acti- excisional biopsy, as GISTs of this size carry a higher risk. vating mutation in either the KIT or PDGFRA (platelet For gastric lesions it is preferable to perform a fine nee - derived growth factor receptor alpha) gene. However, dle aspirate or core needle biopsy under endoscopic ultra- other rare drivers may include mutations in NF1 (neu- sound (EUS) guidance. If this is not feasible, and limited rofibromatosis type 1, loss of function) or BRAF (gain surgery is possible, primary resection may be appropriate. of function). Tumours lacking mutations in KIT or However, the differential diagnosis of intra-abdominal PDGFRA are often called “wild-type”, and those lack- tumours may include leiomyosarcoma, germ cell tumour, ing mutations in not only these genes but also BRAF lymphoma, benign and malignant neurogenic tumours, and NF1 have been dubbed “quadruple wild-type” [8]. and fibromatosis. Given that the management of these Specific advice concerning the management of patients conditions differs substantially, and primary excision is with paediatric, “wild-type” and syndromic GIST can not always appropriate, it is sometimes necessary to do be obtained from a UK based alliance of medical spe- a percutaneous core needle biopsy to confirm the diag - cialists via the web site http://www.pawsgistclinic.org. nosis. This however carries a very small risk of contami - uk. nating the peritoneal cavity, especially if bleeding were to A number of genetic syndromes are linked to GIST: occur. If surgery would require multi visceral resection, or is likely to be morbid, e.g. total gastrectomy, then multiple • The Carney triad syndrome, comprising gastric GIST, core needle biopsies are definitely required, again ideally paraganglioma and pulmonary chondroma (these under EUS guidance, or alternatively using an ultrasound/ may occur at different ages) [9]. computed tomography (CT)-guided percutaneous • Carney–Stratakis syndrome, marked by germ-line approach. Depending on the histological diagnosis initial mutations of one of the SDH subunits A, B, C or D, Judson et al. Clin Sarcoma Res (2017) 7:6 Page 3 of 10 treatment may be with systemic therapy, e.g., lymphomas, sought, so that the tissue is available for later analyses and mesenteric fibromatosis, germ cell tumours and GIST, or research, provided that local ethical approval is in place. alternatively surveillance for benign entities such as non- Consent forms for the National GIST Tissue Bank based at malignant neurogenic tumours. Lesions at risk of rupture, the Royal Marsden Hospital can be found at http://www. such as cystic masses, should only be biopsied in special- givemysample.org/gist. ised centres. If a patient presents with obvious metastatic disease, then a biopsy of an easily accessible metastatic Key recommendations: focus should be performed and a laparotomy for diagnos- 1. Lesions larger than 2  cm in diameter need to be tic purposes is usually unnecessary. investigated and a diagnosis made. Often this can be Pathologically, the diagnosis of GIST relies on morpho- done by endoscopic ultrasound and needle biopsy, logical assessment and immunohistochemistry, the diag- particularly if the lesion is in the stomach. nosis being supported by CD117 immunopositivity [13, 2. The diagnosis should be made by a pathologist expe - 14]. More recently DOG1 has been added to the diagnostic rienced in the disease and include the use of immu- armamentarium [15, 16]. About 5% of GISTs are CD117 nohistochemistry and mutational analysis, which immunonegative, about 5% of GISTs are DOG1 immu- should be performed by an accredited laboratory. nonegative and about 1% of GISTs are immunonegative for 3. If neoadjuvant treatment with imatinib is planned, it both. The mitotic count has prognostic value and although is vital to confirm the diagnosis, since there is a wide several risk assessment tools use an index of mitoses per differential. It may be necessary to perform a percu - 50 high-power fields, it would be more accurate and repro - taneous core needle biopsy if the tumour is inacces- ducible to express this as the number of mitoses in a total sible to endoscopic ultrasound-guided biopsy. Muta- area of 5  mm , which is therefore recommended. If there tional analysis is obligatory, since some GISTs are is some diagnostic doubt, particularly in CD117 and/or insensitive to imatinib (e.g. those with D842V muta- DOG1 immunonegative suspected GIST, molecular analy- tion in exon 18 of PDGFRA). sis for activating mutations in KIT or PDGFRA may help confirm the diagnosis. Mutational analysis has predictive Risk assessment for primary tumours with no value for sensitivity to molecular-targeted therapy, and evidence of metastatic disease also prognostic value. If initial treatment is with imatinib, The TNM classification for staging has several limitations mutational analysis is particularly critical, since some and its use is not recommended in this disease. Prognos- GISTs are insensitive to the drug (e.g. PDGFRA exon 18 tic factors of proven value are the mitotic rate, tumour mutation D842V). The inclusion of mutational analysis in size, tumour site, and presence or absence of tumour the diagnostic work-up of all GISTs should be considered rupture. Gastric GISTs have a better prognosis than small standard practice, with the possible exclusion of sub 2 cm bowel or rectal GISTs. Tumour rupture through a sero- non-rectal GISTs, which are very unlikely ever to need sal surface is an adverse prognostic factor and should be medical treatment. It is strongly recommended that muta- recorded, whether it took place before or during surgery. tional analysis is performed in centralised laboratories Mutational status has not been incorporated in any risk which are enrolled in an external quality assurance pro- classification so far, although some genotypes have a dis - gram, and which have expertise in GIST analysis. In KIT/ tinct natural history, for example KIT/PDGFRA wild type PDGFRA mutation negative, or “Wild-Type” GIST, immu- GISTs tend to exhibit more indolent behaviour than KIT nohistochemistry for succinate dehydrogenase B (SDHB), exon 11 mutant disease. and, if negative, SDHA, should be performed if available, It is possible to assess risk of recurrence after resection since loss of expression may assist the diagnosis and may of a localised GIST. This can be especially useful in decid - help to guide therapy. In the absence of mutations in KIT ing the role of adjuvant therapy in individual patients. or PDGFRA it is also important to look for mutations in Several risk classifications have been proposed. An ini - BRAF, a rare, but important finding from a treatment tial risk classification developed at a consensus meeting perspective, since BRAF inhibitors are available, and also convened by the National Institute of Health (NIH) [13] in NF1. Patients with neurofibromatosis, who have a ger - was useful but only considered size and mitotic index. mline mutation in NF1, are at increased risk of GIST and Although the ‘high-risk’ category based on the NIH cri- this finding may indicate occult neurofibromatosis. The teria has a much worse prognosis than the others, with optimal treatment for advanced GIST with NF1 mutation both ‘very low-risk’ and ‘low-risk’ categories having a has yet to be determined. Collection of fresh/frozen tissue very favourable prognosis, the ‘intermediate-risk’ cat- is encouraged, because new molecular pathology assess- egory did not reliably identify patients with an unfavour- ments can then be made at a later stage in the patient’s able prognosis. A risk classification was proposed by the interest. Informed consent for tumour banking should be Armed Forces Institute of Pathology, which incorporates Judson et al. Clin Sarcoma Res (2017) 7:6 Page 4 of 10 Table 2 Modified NIH risk classification for primary GIST useful, mainly when early assessment of response to tyrosine kinase inhibitor therapy is of special interest, for Risk category Tumour size Mitotic index Primary tumour example after initiation of neo-adjuvant imatinib therapy. (cm) (per 50 HPFs) site Very low risk <2.0 ≤5 Any Treatment Low risk 2.1–5.0 ≤5 Any When small oesophago-gastric or duodenal nodules Intermediate 2.1–5.0 >5 Gastric less than 2  cm in size are detected, endoscopic biopsy risk <5.0 6–10 Any may be difficult and laparoscopic/laparotomic exci - 5.1–10.0 ≤5 Gastric sion may be the only way to make a histological diagno- High risk Any Any Tumour rupture sis. Many of these small nodules, if diagnosed as GIST, >10 Any Any will be low risk, or entities whose clinical significance >5.0 >5 Any remains unclear. Therefore, the standard approach to 2.1-5.0 >5 Non gastric these patients is endoscopic ultrasound assessment, usu- 5.1–10.0 ≤5 Non gastric ally with fine needle aspiration or core needle biopsy, After Joensuu [18] then annual follow-up, reserving excision for patients whose tumour increases in size or becomes symptomatic. Alternatively, the decision about a treatment plan can be the primary tumour site, in addition to mitotic count discussed with the patient. This may depend on age, life and tumour size, i.e. three of the main prognostic fac- expectancy and comorbidities. If follow-up is the choice, tors in localized GISTs [17]. There are also problems definitive evidence regarding an optimal surveillance pol - with the application of this risk classification tool since icy is lacking, but annual follow-up is reasonable. For a mitotic index and tumour size are non-linear continuous small histologically proven GIST of 2  cm or greater, the variables, so that the risk thresholds, especially based on standard treatment is excision, unless major morbidity number of mitoses, need to be interpreted with care. A is expected. Alternatively, in the case of a low-risk GIST, modification to the NIH criteria has been proposed that the possibility of surveillance could be discussed with the incorporates tumour site and additionally tumour rup- patient. Rectal (or recto-vaginal space) nodules should be ture, an important risk factor, [18] see Table 2. Prognostic biopsied and preferably excised after ultrasound assess- contour maps have been generated through a number of ment, regardless of tumour size. This is because GISTs at series of GIST patients not treated with adjuvant therapy. these sites have a higher risk of local recurrence after sur- These maps incorporate the mitotic index and tumour gery, and the local implications for surgery in relation to size as continuous non-linear variables, while tumour morbidity are more critical. In specific clinical contexts, rupture is considered in addition to tumour site. They if the tumour is small, a follow-up policy without surgery have been validated against pooled data from 10 series could be adopted, but such an approach should be dis- and 2560 patients from the literature [19]. Several nomo- cussed in detail with the patient. grams and web or mobile phone applications are avail- Multidisciplinary treatment planning is needed. This able to enable rapid risk category calculations to be made, should involve histopathologists, radiologists, surgeons, which may assist multidisciplinary planning of patient and oncologists, as well as gastroenterologists, nuclear management. medicine specialists and others, as applicable. Such teams are available in reference centres for sarcomas Staging procedures and GISTs, which treat a large number of GIST patients Staging procedures take into account the fact that most annually. Support staff, such as clinical nurse specialists, relapses affect the peritoneum and the liver. Contrast- play a vital role and are not likely to be available, or have enhanced abdominal and pelvic CT scan is the inves- the appropriate expertise, outside specialist centres. tigation of choice for staging and follow-up. Magnetic resonance imaging (MRI) or contrast-enhanced ultra- Localised disease—surgery sound may be alternatives, especially in younger patients Standard treatment of localized GISTs is complete surgi- where exposure to radiation should be limited. MRI pro- cal excision of the lesion, with no dissection of clinically vides better preoperative staging information for rectal negative lymph nodes [III, A]. Surgery should be per- GISTs. Chest CT scan or X-ray and routine laboratory formed by a sub speciality surgeon who is fully trained testing complement the staging work-up of the asymp- and experienced in radical anatomic site specific can - tomatic patient, but are not routinely required during cer surgery. When adjacent organs are involved, en bloc follow-up. Evaluation of F-fluorodeoxyglucose (FDG) resection is recommended wherever possible. If laparo- uptake using an FDG-positron emission tomography scopic excision is planned, the technique needs to follow (PET) scan, or FDG-PET–CT/MRI, can sometimes be Judson et al. Clin Sarcoma Res (2017) 7:6 Page 5 of 10 the principles of oncological surgery [20] [III, A]. A lap- and was approved by the National Institute for Care and aroscopic approach is clearly discouraged in patients Health Excellence (NICE) in their recent re-appraisal with large tumours, because of the risk of tumour rup- (https://www.nice.org.uk/guidance/ta326). Mutational ture, which is associated with a very high risk of relapse. analysis is critical to making a clinical decision regard- R0 excision is the goal (i.e., an excision whose margins ing adjuvant therapy. In fact, there is a consensus that are clear of tumour cells). When R0 surgery is likely to PDGFRA D842  V-mutated GISTs should not be treated result in major functional sequelae, e.g. total gastrec- with any adjuvant therapy, given the lack of sensitivity of tomy or abdomino-perineal resection of rectum, neo- this genotype to imatinib both in  vitro and in  vivo [IV, adjuvant imatinib should be regarded as standard therapy A]. Given the data supporting the use of a higher dose of [21–24] [IV, A]. Treatment is given to reduce the size imatinib (800 mg daily) in the presence of an exon 9 KIT of the tumour and to limit subsequent surgical mor- mutation in advanced GIST, clinicians might consider bidity. This may also be the case if surgery will be safer using this dose in the adjuvant setting for this genotype following cyto-reduction (e.g. the risk of bleeding and [28–31]. However, this is not supported by any controlled tumour rupture is likely to be decreased). After maxi- trial data in the adjuvant setting and use of the higher mal tumour response, generally after 6–12  months, sur- dose is not approved by NICE in the UK. There is con - gery is performed. Prior mutational analysis is crucial to sensus on avoiding adjuvant treatment in NF-1 related prevent patients with less sensitive or resistant tumours GISTs, which are insensitive to imatinib in the advanced (e.g. PDGFRA D842  V mutations) from receiving ther- setting. On the other hand, a consensus is lacking among apy with imatinib, and to allow appropriate dosing for experts about whether KIT/PDGFRA wild-type SDH- patients with KIT exon 9 mutated tumours. Early tumour negative GIST should be treated with adjuvant therapy. response assessment is mandatory, so that surgery is not This reflects their lower sensitivity to imatinib, as well as delayed in the case of non-responding disease. Functional their peculiar natural history, which is often more indo- imaging, such as PET-CT, makes it possible to assess the lent; subgroup analyses of available randomized trials are tumour response very rapidly, within a few weeks. There too limited to provide sufficient evidence. are limited data to guide the physician on when to stop If there has been tumour rupture before or during sur- imatinib before surgery, but it can safely be stopped a few gery, there will have been spillage of tumour cells into the days, or even 1 day, before surgery and it can be resumed peritoneal cavity, and therefore occult peritoneal disease promptly when the patient has recovered from the acute can be assumed to exist. This puts the patient at a very effects of surgery. high risk of peritoneal relapse. Therefore, these patients If preoperative medical treatment has not helped or should be considered for adjuvant imatinib therapy. The cannot be used, there should be a discussion with the optimal duration of treatment in these cases is unknown, patient about accepting a possible R1 resection with given the uncertainty as to whether they should be microscopically positive margins (i.e. excision mar- viewed as essentially having metastatic disease, but gins containing tumour cells) [IV, B]. This may be more should be at least 3 years, as for high risk resected GIST. acceptable for low-risk lesions, with the lack of any for- mal demonstration that R1 surgery is associated with Key recommendations worse overall survival [25]. If an unplanned R1 exci- 1. GIST should be managed by an experienced multi- sion has already been carried out, re-excision may be an disciplinary team. option, provided the original site of lesion can be found, 2. Pre-operative imatinib should be considered for and major functional sequelae are not foreseen. those large gastric or rectal primaries where immedi- ate resection is likely to be morbid, e.g. total gastrec- Localised disease—adjuvant therapy tomy or abdomino-perineal resection. In this situ- The risk of relapse following surgery can be substan - ation mutational analysis is mandatory prior to the tial, as defined by available risk classifications. Adjuvant initiation of imatinib therapy. treatment with imatinib for 3  years was associated with 3. Patients at high risk of recurrence or distant relapse improved relapse-free and overall survival compared should receive 3 years of adjuvant imatinib, provided with 1  year of therapy in a randomized trial in high-risk their tumour is not likely to be resistant to therapy patients [26]. Previously, a placebo-controlled trial dem- (PDGFRA exon 18 mutation D842V). onstrated that imatinib given for 1 year prolongs relapse- free survival in localized GISTs larger than 3  cm with a Metastatic disease—systemic treatment macroscopically complete resection [27]. Therefore, adju - In patients with inoperable and metastatic disease, vant therapy with imatinib for 3  years is standard treat- imatinib is the standard treatment [32, 33] [III, A], includ- ment for patients with a significant risk of relapse [I, A] ing patients who had previously received the drug as Judson et al. Clin Sarcoma Res (2017) 7:6 Page 6 of 10 adjuvant therapy without relapse during this treatment. shown to be related to a good prognosis, provided the This also applies to metastatic patients whose disease has patient is responding to imatinib, but whether this is due been completely removed surgically, although surgery to surgery or to patient selection [38–40] has never been as a primary approach to metastatic GIST is not recom- demonstrated prospectively. Conducting a randomised mended. The standard dose of imatinib is 400  mg daily trial did not prove feasible; thus, at the present time sur- [I, A]. However, data have shown that patients with KIT gery can be discussed with the patient but not recom- exon 9 mutations fare better in terms of progression-free mended on the basis of a definitive proof of benefit [III, survival (PFS) on a higher dose of 800 mg daily, which is C]. Surgical excision of progressive disease is not recom- therefore the standard treatment in this subgroup [31] mended, given the poor results in published series, but [III, A], albeit not recommended by a NICE appraisal surgery of limited progression, such as the ‘nodule within which only assessed dose escalation in the context of dis- a mass’, has been associated with a progression-free inter- ease progression. A report on the long term follow-up of val in the same range as for second-line treatment with the European/Australasian clinical trial which compared sunitinib. So, this may be a palliative option in the indi- 400  mg with 800  mg imatinib in patients with advanced vidual patient with limited progression, while continuing GIST has shown a survival advantage for the initial use imatinib [V, C]. Non-surgical procedures, such as radi- of the 800 mg dose in those with exon 9 mutations in KIT ofrequency ablation of liver metastases may also be used. (Casali P et al., in press) indicating a need to review this Prior to performing such interventions PET-CT can issue in the UK. Treatment should be continued indefi - be useful to confirm the location of imatinib-resistant nitely, since treatment interruption is generally followed disease. by relatively rapid tumour progression, even when lesions Dose escalation of imatinib to 800  mg in the case of a have previously been surgically excised [34] [II, B]. At GIST with a KIT exon 9 mutation showing disease pro- the start of treatment the patient should be alerted to gression could be considered if the higher dose was not the importance of adherence to therapy, and of possible used initially, since the higher dose is significantly more interactions with concomitant medications and foods. effective in this setting [31]. Higher doses, though not They should also be given guidance about the best ways necessarily 800 mg, could be useful if satisfactory plasma to handle any possible side effects. Dose intensity should levels of imatinib are not being achieved, but the use of be maintained by effective management of side effects, higher doses is not approved by NICE. The potential and a rational policy of dose reductions and interruptions misinterpretation of the images produced by the com- should be applied if there is excessive, persistent toxicity. plex tumour response patterns to TKIs can lead to a false Retrospective data suggest that suboptimal plasma levels diagnosis of progression, which must be considered. of imatinib are associated with a worse outcome, though Patient non-compliance and drug interactions with con- a correlation with outcome has not been established pro- comitant medications must also be ruled out as the pos- spectively [35]. A recent report confirmed that patients sible cause of progression. with imatinib trough levels of less than 760 ng/ml, taken If there is confirmed progression, or rare intolerance after a minimum of 3  months’ treatment, which equates to imatinib after all attempts to manage side effects have to steady state [36], had a worse outlook in terms of pro- failed, the standard second-line treatment is the tyros- gression-free survival, which applied in the case of both ine kinase inhibitor (TKI) sunitinib [41] [I, B]. This drug gastric and small bowel GIST [37]. Aside from its poten- was proven to be effective in terms of PFS using a regi - tial use to tailor the imatinib dose, plasma level assess- men of 50  mg daily 4  weeks on/2  weeks off. Data have ment may be useful in the case of: (i) patients receiving been published showing that continuous treatment with concomitant medications that put them at a risk of major a lower daily dose of 37.5  mg is also effective and well interactions; (ii) unexpected observed toxicities; (iii) tolerated, although no formal comparison has been per- progression on 400  mg. Dose adaptation according to formed within a randomized clinical trial. This schedule inadequate imatinib trough level is being studied in the can therefore be considered an alternative on an individ- Netherlands, and is a standard approach in a number of ualized basis [42] [III, B]. However, not all patients resist- institutions. However, the use of a higher dose of imatinib ant to imatinib respond to sunitinib particularly those in patients with progressive disease is not approved by with secondary mutations affecting the activation loop NICE. domain of KIT and the PDGFRA exon 18 D842V muta- Close monitoring of the tumour response should be tion, which is always resistant. carried out in the early phases of treatment. Follow-up A prospective placebo-controlled randomized trial should be continued throughout the treatment, since the demonstrated that regorafenib, at a dose of 160 mg daily risk of secondary progression persists over time. Com- on a 3  weeks on/1  week off schedule, significantly pro - plete excision of residual metastatic disease has been longed PFS in patients progressing after both imatinib Judson et al. Clin Sarcoma Res (2017) 7:6 Page 7 of 10 and sunitinib [43]. Regorafenib is regarded as standard Response assessment therapy for the third-line treatment of patients progress- Response assessment is complex and early progression ing on or failing to respond to imatinib and sunitinib [I, in particular should be confirmed by a team experienced B]. It is currently available in England via the National in treating GIST. Anti-tumour activity translates into Cancer Drugs Fund and is also available in Scotland, tumour shrinkage in most patients, but some patients Wales and Northern Ireland as standard 3rd line therapy. may show only changes in tumour “density” on imaging, The key distinction between sunitinib and regorafenib, these changes sometimes precede a reduction in tumour as also previously shown with the analogue sorafenib, is volume. Such changes in tumour radiological appearance its ability to inhibit tumours with secondary mutations in should be considered as indicative of tumour response. the activation loop of KIT, especially in exon 17 [44, 45]. Tumour size may even increase in the short term but if These mutations are known to confer resistance both to tumour density on CT scan is decreased this may still imatinib and sunitinib, hence the value of regorafenib in indicate tumour response [48, 49]. Even the apparent this setting. ‘appearance’ of new lesions may be due to them becom- Patients with metastatic GIST failing all three standard ing less dense, or cystic, especially in the liver. Therefore, agents should be considered for participation in clinical both tumour size and tumour density on CT scan, or trials of new agents. These studies are only likely to be consistent changes on MRI or contrast-enhanced ultra- available in major centres treating GIST. There is limited sound, should be considered when determining tumour evidence that patients who have already progressed on response. F-FDG-PET has proved useful in the early imatinib may benefit for a limited period when re-chal - assessment of tumour response, for example when pre- lenged with the drug [46]. Likewise, there is anecdotal diction of the response is valuable, for example in the case evidence that maintaining treatment with a TKI even in of preoperative therapy, or when response is in doubt. the case of progressive disease, as opposed to stopping it, However, a small proportion of GISTs have no FDG may slow down progression if no other option is available uptake. The absence of tumour progression at 6  months at the time. Therefore, re-challenging or continuing treat - [50] is also equivalent to a tumour response. Conversely, ment with a TKI, to which the patient has already been tumour progression may not always be accompanied by exposed, is an option which may be considered for symp- changes in tumour size. For example, an increase in the tom control in patients with progression [V, B]. tumour density shown by contrast enhancement within a previously responding low density tumour lesion, may be Metastatic disease—local therapy indicative of tumour progression. A typical progression Selected patients with limited liver metastatic disease pattern is the ‘nodule within the mass’, in which a portion may be amenable to surgery or radiofrequency ablation of a responding lesion becomes hyper-dense [51]. (RFA) after maximum response to imatinib, or if there is evidence of localised disease progression. The use of RFA Key recommendations is restricted to tumours in the region of 3  cm in maxi- 1. Imatinib is the treatment of choice for patients with mum diameter and is less likely to be a suitable approach unresectable or metastatic disease and is given until for lesions adjacent to large vessels or superficial lesions, progression at the standard dose of 400  mg daily. especially if displacing the liver capsule. However, larger Data suggest that patients whose tumours have an isolated lesions and superficial lesions may still be suit - exon 9 mutation in KIT benefit from a larger dose, able for surgical resection, either by partial hepatectomy though this is not currently recommended by NICE. or wedge resection. Dedicated liver MRI scans and when 2. Isolated progression may be amenable to surgery or appropriate PET-CT scans may be required to determine other local measures, such as radiofrequency abla- whether this is a legitimate approach by excluding other tion. occult active disease. 3. Standard second line treatment is sunitinib, which Radiotherapy can be a useful local therapy in GIST may be given at the recommended dose of 50  mg under certain circumstances in the advanced disease set- daily for 4 weeks every 6 weeks, or 37.5 mg daily con- ting. If there is a single site of disease that is progressing tinuously. on a TKI and can be encompassed within a radiotherapy 4. Standard 3rd line treatment is regorafenib. treatment field, then radiotherapy delivered to a moder - ate or high dose can offer local tumour control, and pos - Follow‑up sibly prolong the use of the TKI [47]. Radiotherapy can The optimal follow-up policy for surgically treated also be used at lower doses to palliate symptomatic dis- patients with localized disease is unclear. Relapses occur ease, for example to relieve pain or bleeding. most often in the liver and/or the peritoneal cavity. Other sites of metastases, including bone and brain are Judson et al. Clin Sarcoma Res (2017) 7:6 Page 8 of 10 Authors’ contributions uncommon, but may be less unusual following prolonged All authors contributed to the content of the manuscript and consent to its treatment with several lines of therapy. The mitotic rate publication. IJ adapted the guideline from the previous version published in most likely affects the frequency with which relapses 2010, with reference to the ESMO Guidelines from 2014 of which he was a co-author. Amendments and suggestions from the co-authors were incorpo- occur. Risk assessment based on the mitotic count, rated. All authors read and approved the final manuscript. tumour size and tumour site may be useful in choosing the routine follow-up policy. High-risk patients generally Author details The Institute of Cancer Research, Royal Marsden NHS Foundation Trust, relapse within 1–3  years from the end of adjuvant ther- Fulham Road, London SW3 6JJ, UK. Addenbrooke’s Hospital, Cambridge apy. Low-risk patients may relapse later, given that the University Hospitals, Cambridge, UK. University College Hospital, University disease is likely to be slower growing. College London Hospitals NHS Foundation Trust, London, UK. Bristol Cancer Institute, University Hospitals, Bristol NHS Trust, Bristol, UK. Southmead The issue of follow-up has been addressed by Joen - Hospital, North Bristol NHS Trust, Bristol, UK. Royal Marsden NHS Foundation suu and colleagues based on the currently available Trust London, London, UK. evidence [52]. High-risk patients who have undergone Acknowledgements resection of their primary generally undergo a rou- The authors would also like to thank the following on behalf of the British tine follow-up with abdominal CT or MRI scan every Sarcoma Group for valuable review, suggestions and amendments: Dr. Robin 3–6  months during adjuvant therapy, for 3  years. This Jones (Consultant Medical Oncologist, Royal Marsden Hospital, London), Dr. Palma Dileo (Consultant Medical Oncologist, University College Hospital, frequency of follow up is because of the need to man- London), Mrs. Judith Robinson (GIST Support UK), Jayne Bressington (GIST age the side effects of the therapy. On cessation of adju - Support UK), Dr. Nicholas Carroll (Consultant Radiologist, Addenbrooke’s vant therapy follow up is every 3  months for 2  years, Hospital, Cambridge). then every 6  months for another 3  years, after which Competing interests follow up is annually for another 5 years. Patients with IJ has received honoraria for attending advisory boards and speaking engage- high-risk tumours not given adjuvant therapy, for ments from GSK, Ariad, Amgen, Clinigen, Lilly and Bayer; BS has received honoraria for attending advisory boards from Ariad, Novartis, Clinigen, Lilly whatever reason, should be followed up 3 monthly for and Daichi. NW has received honoraria from Novartis for speaking engage- 2  years, 6 monthly for 3  years and then annually for a ments; RB has received honoraria for attending advisory boards and speaking further 5 years. engagements from Novartis, Pfizer, Ariad, Pierre Fabre, Bristol Myers Squibb, Astra Zeneca, and Bayer Pharmaceuticals. AD & SM have no competing For low to intermediate risk tumours, the optimal interests to report. frequency of follow-up is less clear. If follow-up is per- formed, it will usually be an abdominal CT or MRI scan, Availability of data No clinical or biological data was included in the preparation of these or ultrasound, every 6–12 months for 5 years. guidelines. Very low-risk GISTs do not require routine follow-up, provided excision was complete, although one must be Ethics approval and consent to participate These guidelines do not contain information on patients, hence neither ethi- aware that the risk of progression is not zero. cal approval nor consent from patients was required. Radiation exposure is a factor to consider when select- ing the imaging modality for long term follow-up. Publisher’s Note Abdominal MRI is an acceptable alternative to CT which Springer Nature remains neutral with regard to jurisdictional claims in pub- could be used at selected intervals. lished maps and institutional affiliations. Received: 2 February 2017 Accepted: 6 April 2017 Key recommendations 1. Patients with high risk disease on adjuvant therapy should be followed up by cross-sectional imaging every 3–6 months during their 3 years of treatment, 3 monthly for 2  years following cessation of treat- References 1. Group ESESNW. Gastrointestinal stromal tumours: ESMO clinical practice ment and thereafter every 6  months for 3  years and guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2014;25 annually for 5 years. Suppl 3:iii21–6. 2. Patients with high risk disease not receiving adjuvant 2. Dykewicz CA, Centers for Disease C, Prevention, Infectious Diseases Society of A, American Society of B, Marrow T. Summary of the guidelines treatment should follow the post-adjuvant treatment for preventing opportunistic infections among hematopoietic stem cell scheme. transplant recipients. Clin Infect Dis. 2001;33(2):139–44. 3. For patients with lesser levels of risk less frequent fol- 3. Gatta G, van der Zwan JM, Casali PG, Siesling S, Dei Tos AP, Kunkler I, et al. Rare cancers are not so rare: the rare cancer burden in Europe. Eur J low-up is generally recommended, although the clini- Cancer. 2011;47(17):2493–511. cal benefits are unclear. For intermediate risk patients 4. Cassier PA, Ducimetiere F, Lurkin A, Ranchere-Vince D, Scoazec JY, Bringuier 6 monthly scans for 5 years followed by annual scans PP, et al. A prospective epidemiological study of new incident GISTs during two consecutive years in Rhone Alpes region: incidence and molecular and for low risk patients shorter duration follow-up distribution of GIST in a European region. Br J Cancer. 2010;103(2):165–70. would be considered reasonable. 5. Pappo AS, Janeway KA. Pediatric gastrointestinal stromal tumors. Hema- tol Oncol Clin N Am. 2009;23(1):15–34. Judson et al. Clin Sarcoma Res (2017) 7:6 Page 9 of 10 6. Agaimy A, Wunsch PH, Sobin LH, Lasota J, Miettinen M. Occurrence of 27. Dematteo RP, Ballman KV, Antonescu CR, Maki RG, Pisters PW, Demetri other malignancies in patients with gastrointestinal stromal tumors. GD, et al. Adjuvant imatinib mesylate after resection of localised, primary Semin Diagn Pathol. 2006;23(2):120–9. gastrointestinal stromal tumour: a randomised, double-blind, placebo- 7. Ballinger ML, Goode DL, Ray-Coquard I, James PA, Mitchell G, Niedermayr controlled trial. Lancet. 2009;373(9669):1097–104. E, et al. Monogenic and polygenic determinants of sarcoma risk: an 28. Debiec-Rychter M, Sciot R, Le Cesne A, Schlemmer M, Hohenberger P, international genetic study. Lancet Oncol. 2016;17:1261–71. van Oosterom AT, et al. KIT mutations and dose selection for imatinib in 8. Pantaleo MA, Nannini M, Corless CL, Heinrich MC. Quadruple wild-type patients with advanced gastrointestinal stromal tumours. Eur J Cancer. ( WT ) GIST: defining the subset of GIST that lacks abnormalities of KIT, 2006;42(8):1093–103. PDGFRA, SDH, or RAS signaling pathways. Cancer Med. 2015;4(1):101–3. 29. Heinrich MC, Owzar K, Corless CL, Hollis D, Borden EC, Fletcher CD, 9. Zhang L, Smyrk TC, Young WF Jr, Stratakis CA, Carney JA. Gastric stromal et al. Correlation of kinase genotype and clinical outcome in the North tumors in Carney triad are different clinically, pathologically, and behavio - American Intergroup Phase III Trial of imatinib mesylate for treatment of rally from sporadic gastric gastrointestinal stromal tumors: findings in 104 advanced gastrointestinal stromal tumor: CALGB 150105 Study by Cancer cases. Am J Surg Pathol. 2010;34(1):53–64. and Leukemia Group B and Southwest Oncology Group. J Clin Oncol. 10. Pasini B, McWhinney SR, Bei T, Matyakhina L, Stergiopoulos S, Muchow 2008;26(33):5360–7. M, et al. Clinical and molecular genetics of patients with the Carney– 30. Heinrich MC, Corless CL, Demetri GD, Blanke CD, von Mehren M, Joensuu Stratakis syndrome and germline mutations of the genes coding for the H, et al. Kinase mutations and imatinib response in patients with meta- succinate dehydrogenase subunits SDHB, SDHC, and SDHD. Eur J Hum static gastrointestinal stromal tumor. J Clin Oncol. 2003;21(23):4342–9. Genet. 2008;16(1):79–88. 31. Gastrointestinal Stromal Tumor Meta-Analysis G. Comparison of two 11. Gaal J, Stratakis CA, Carney JA, Ball ER, Korpershoek E, Lodish MB, et al. doses of imatinib for the treatment of unresectable or metastatic gastro- SDHB immunohistochemistry: a useful tool in the diagnosis of Carney– intestinal stromal tumors: a meta-analysis of 1640 patients. J Clin Oncol. Stratakis and Carney triad gastrointestinal stromal tumors. Mod Pathol. 2010;28(7):1247–53. 2011;24(1):147–51. 32. Verweij J, Casali PG, Zalcberg J, LeCesne A, Reichardt P, Blay JY, et al. 12. Miettinen M, Fetsch JF, Sobin LH, Lasota J. Gastrointestinal stromal tumors Progression-free survival in gastrointestinal stromal tumours with high- in patients with neurofibromatosis 1: a clinicopathologic and molecular dose imatinib: randomised trial. Lancet. 2004;364(9440):1127–34. genetic study of 45 cases. Am J Surg Pathol. 2006;30(1):90–6. 33. Blanke CD, Rankin C, Demetri GD, Ryan CW, von Mehren M, Benjamin RS, 13. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, et al. et al. Phase III randomized, intergroup trial assessing imatinib mesylate at Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum two dose levels in patients with unresectable or metastatic gastrointes- Pathol. 2002;33(5):459–65. tinal stromal tumors expressing the kit receptor tyrosine kinase: S0033. J 14. Rubin BP, Cooper K, Fletcher CD, Folpe AL, Gannon FH, Hunt JL, et al. Clin Oncol. 2008;26(4):626–32. Protocol for the examination of specimens from patients with tumors of 34. Le Cesne A, Ray-Coquard I, Bui BN, Adenis A, Rios M, Bertucci F, et al. soft tissue. Arch Pathol Lab Med. 2010;134(4):e31–9. Discontinuation of imatinib in patients with advanced gastrointestinal 15. Miettinen M, Wang ZF, Lasota J. DOG1 antibody in the differential diag- stromal tumours after 3 years of treatment: an open-label multicentre nosis of gastrointestinal stromal tumors: a study of 1840 cases. Am J Surg randomised phase 3 trial. Lancet Oncol. 2010;11(10):942–9. Pathol. 2009;33(9):1401–8. 35. Demetri GD, Wang Y, Wehrle E, Racine A, Nikolova Z, Blanke CD, et al. 16. Lopes LF, West RB, Bacchi LM, van de Rijn M, Bacchi CE. DOG1 for Imatinib plasma levels are correlated with clinical benefit in patients with the diagnosis of gastrointestinal stromal tumor (GIST ): comparison unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol. between 2 different antibodies. Appl Immunohistochem Mol Morphol. 2009;27(19):3141–7. 2010;18(4):333–7. 36. Eechoute KFM, Reyners AK, de Jong FA, Sparreboom A, van der Graaf 17. Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and WTA, et al. A long-term prospective population pharmacokinetic study prognosis at different sites. Semin Diagn Pathol. 2006;23(2):70–83. on imatinib plasma concentrations in GIST patients. Clin Cancer Res. 18. Joensuu H. Risk stratification of patients diagnosed with gastrointestinal 2012;18:5780–7. stromal tumor. Hum Pathol. 2008;39(10):1411–9. 37. Bouchet S, Poulette S, Titier K, Moore N, Lassalle R, Abouelfath A, et al. 19. Joensuu H, Vehtari A, Riihimaki J, Nishida T, Steigen SE, Brabec P, et al. Risk Relationship between imatinib trough concentration and outcomes in of recurrence of gastrointestinal stromal tumour after surgery: an analysis the treatment of advanced gastrointestinal stromal tumours in a real-life of pooled population-based cohorts. Lancet Oncol. 2012;13(3):265–74. setting. Eur J Cancer. 2016;57:31–8. 20. Novitsky YW, Kercher KW, Sing RF, Heniford BT. Long-term outcomes of 38. Raut CP, Posner M, Desai J, Morgan JA, George S, Zahrieh D, et al. Surgical laparoscopic resection of gastric gastrointestinal stromal tumors. Ann management of advanced gastrointestinal stromal tumors after treat- Surg. 2006;243(6):738–45 (discussion 45–7). ment with targeted systemic therapy using kinase inhibitors. J Clin Oncol. 21. Eisenberg BL, Harris J, Blanke CD, Demetri GD, Heinrich MC, Watson JC, 2006;24(15):2325–31. et al. Phase II trial of neoadjuvant/adjuvant imatinib mesylate (IM) for 39. Gronchi A, Fiore M, Miselli F, Lagonigro MS, Coco P, Messina A, et al. advanced primary and metastatic/recurrent operable gastrointestinal Surgery of residual disease following molecular-targeted therapy stromal tumor (GIST ): early results of RTOG 0132/ACRIN 6665. J Surg with imatinib mesylate in advanced/metastatic GIST. Ann Surg. Oncol. 2009;99(1):42–7. 2007;245(3):341–6. 22. Cananzi FC, Judson I, Lorenzi B, Benson C, Mudan S. Multidisciplinary care 40. Mussi C, Ronellenfitsch U, Jakob J, Tamborini E, Reichardt P, Casali PG, of gastrointestinal stromal tumour: a review and a proposal for a pre- et al. Post-imatinib surgery in advanced/metastatic GIST: is it worthwhile treatment classification. Eur J Surg Oncol. 2013;39(11):1171–8. in all patients? Ann Oncol. 2010;21(2):403–8. 23. Wilkinson MJ, Fitzgerald JE, Strauss DC, Hayes AJ, Thomas JM, Messiou C, 41. Demetri GD, van Oosterom AT, Garrett CR, Blackstein ME, Shah MH, et al. Surgical treatment of gastrointestinal stromal tumour of the rectum Verweij J, et al. Efficacy and safety of sunitinib in patients with advanced in the era of imatinib. Br J Surg. 2015;102(8):965–71. gastrointestinal stromal tumour after failure of imatinib: a randomised 24. Rutkowski P, Gronchi A, Hohenberger P, Bonvalot S, Schoffski P, Bauer controlled trial. Lancet. 2006;368(9544):1329–38. S, et al. Neoadjuvant imatinib in locally advanced gastrointestinal 42. George S, Blay JY, Casali PG, Le Cesne A, Stephenson P, Deprimo SE, et al. stromal tumors (GIST ): the EORTC STBSG experience. Ann Surg Oncol. Clinical evaluation of continuous daily dosing of sunitinib malate in 2013;20(9):2937–43. patients with advanced gastrointestinal stromal tumour after imatinib 25. McCarter MD, Antonescu CR, Ballman KV, Maki RG, Pisters PW, Demetri failure. Eur J Cancer. 2009;45(11):1959–68. GD, et al. Microscopically positive margins for primary gastrointestinal 43. Demetri GD, Reichardt P, Kang YK, Blay JY, Rutkowski P, Gelderblom H, stromal tumors: analysis of risk factors and tumor recurrence. J Am Coll et al. Efficacy and safety of regorafenib for advanced gastrointestinal Surg. 2012;215(1):53–9 (discussion 9–60). stromal tumours after failure of imatinib and sunitinib (GRID): an interna- 26. Joensuu H, Eriksson M, Sundby Hall K, Hartmann JT, Pink D, Schutte J, tional, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet. et al. One vs three years of adjuvant imatinib for operable gastrointestinal 2013;381(9863):295–302. stromal tumor: a randomized trial. JAMA. 2012;307(12):1265–72. 44. Heinrich MC, Marino-Enriquez A, Presnell A, Donsky RS, Griffith DJ, McKinley A, et al. Sorafenib inhibits many kinase mutations associated Judson et al. Clin Sarcoma Res (2017) 7:6 Page 10 of 10 with drug-resistant gastrointestinal stromal tumors. Mol Cancer Ther. tomography in patients with metastatic gastrointestinal stromal 2012;11(8):1770–80. tumor treated at a single institution with imatinib mesylate: pro- 45. Hsueh YS, Lin CL, Chiang NJ, Yen CC, Li CF, Shan YS, et al. Selecting tyros- posal of new computed tomography response criteria. J Clin Oncol. ine kinase inhibitors for gastrointestinal stromal tumor with secondary 2007;25(13):1753–9. KIT activation-loop domain mutations. PLoS ONE. 2013;8(6):e65762. 50. Le Cesne A, Van Glabbeke M, Verweij J, Casali PG, Findlay M, Reichardt P, 46. Kang YK, Ryu MH, Yoo C, Ryoo BY, Kim HJ, Lee JJ, et al. Resumption of et al. Absence of progression as assessed by response evaluation criteria imatinib to control metastatic or unresectable gastrointestinal stromal in solid tumors predicts survival in advanced GI stromal tumors treated tumours after failure of imatinib and sunitinib (RIGHT ): a randomised, with imatinib mesylate: the intergroup EORTC-ISG-AGITG phase III trial. J placebo-controlled, phase 3 trial. Lancet Oncol. 2013;14(12):1175–82. Clin Oncol. 2009;27(24):3969–74. 47. Joensuu H, Eriksson M, Collan J, Balk MH, Leyvraz S, Montemurro M. 51. Shankar S, vanSonnenberg E, Desai J, Dipiro PJ, Van Den Abbeele A, Radiotherapy for GIST progressing during or after tyrosine kinase inhibi- Demetri GD. Gastrointestinal stromal tumor: new nodule-within-a-mass tor therapy: a prospective study. Radiother Oncol. 2015;116(2):233–8. pattern of recurrence after partial response to imatinib mesylate. Radiol- 48. Benjamin RS, Choi H, Macapinlac HA, Burgess MA, Patel SR, Chen ogy. 2005;235(3):892–8. LL, et al. We should desist using RECIST, at least in GIST. J Clin Oncol. 52. Joensuu H, Martin-Broto J, Nishida T, Reichardt P, Schoffski P, Maki RG. 2007;25(13):1760–4. Follow-up strategies for patients with gastrointestinal stromal tumour 49. Choi H, Charnsangavej C, Faria SC, Macapinlac HA, Burgess MA, Patel treated with or without adjuvant imatinib after surgery. Eur J Cancer. SR, et al. Correlation of computed tomography and positron emission 2015;51(12):1611–7. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Clinical Sarcoma Research Springer Journals

UK clinical practice guidelines for the management of gastrointestinal stromal tumours (GIST)

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/uk-clinical-practice-guidelines-for-the-management-of-gastrointestinal-gfy7QZOWa4

References (111)

Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology
eISSN
2045-3329
DOI
10.1186/s13569-017-0072-8
Publisher site
See Article on Publisher Site

Abstract

Background: Soft tissue sarcomas (STS) are rare tumours arising in mesenchymal tissues. Gastrointestinal stromal tumour (GIST ) is the commonest STS and arises within the wall of the gastrointestinal (GI) tract. While most GISTs occur in the stomach they do occur in all parts of the GI tract. As with other STS, it is important that GISTs are man- aged by expert teams, to ensure consistent and optimal treatment, as well as recruitment to clinical trials, and the ongoing accumulation of further knowledge of the disease. The development of appropriate guidance, by an expe- rienced panel referring to the evidence available, is therefore a useful foundation on which to build progress in the field. Methodology: British Sarcoma Group guidelines for the management of GIST were initially developed by a panel of physicians experienced in the management of GIST. This current version has been updated and amended with refer- ence to other European and US guidance. We have received input from representatives of all diagnostic and treat- ment disciplines as well as patient representatives. Levels of evidence and strength of recommendation gradings are those used by ESMO adapted from those published by the Infectious Disease Society of America. Conclusions: The guidelines cover aetiology, genetics and underlying molecular mechanisms, diagnosis and initial investigations, staging and risk stratification, surgery, neoadjuvant and adjuvant therapy, the management of advanced disease and follow-up. The importance of mutational analysis in guiding treatment is highlighted, since this can indicate the most effective treatment and avoid administration of ineffective drugs, emphasising the need for management in specialist centres. Background Incidence British Sarcoma Group guidelines for the management Gastrointestinal stromal tumours (GISTs) are rare of GIST were initially developed in 2004 by a panel of cancers, with an estimated unadjusted incidence of physicians experienced in the management of GIST rep- 1.5/100,000/year [3]. Data from the Rhȏne-Alpes region resenting all diagnostic and treatment disciplines. These of France [4] and “NHS England Cancer Registry” (per- Clinical Practice Guidelines update reflect the improve - sonal communication) suggest an incidence of just under ments in our knowledge of the disease and developments 11 per million per annum, equating to 650 clinically that have taken place since then [1]. Levels of evidence meaningful new cases a year in the UK, approximately and strength of recommendation gradings are those used 900 in total. Accurate data on prevalence in the UK are by ESMO adapted from those published by the Infectious not yet available. Disease Society of America (Table 1) [2]. The median age at diagnosis is around 60–65  years, with a wide range. Occurrence in children, adolescents and younger patients is very rare, although paediatric *Correspondence: Ian.Judson@icr.ac.uk GISTs represent a distinct subset, marked by female pre- The Institute of Cancer Research, Royal Marsden NHS Foundation Trust, dominance, absence of KIT/platelet-derived growth fac- Fulham Road, London SW3 6JJ, UK Full list of author information is available at the end of the article tor alpha (PDGFRA) mutations, usually gastric origin, © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Judson et al. Clin Sarcoma Res (2017) 7:6 Page 2 of 10 Table 1 Levels of evidence and Grades of recommendation leading to a dyad of GIST and paraganglioma [10, 11]. Levels of evidence • Type-1 neurofibromatosis, i.e. associated with loss of I Evidence from at least one large randomised, controlled trial of good function of NF1, whether sporadic or inherited, and methodological quality (low potential for a bias) or meta-analyses of well-conducted randomised trials without heterogeneity absence of mutations in KIT or PDGFRA, the GISTs are often multifocal, predominantly located in the II Small randomised trials or large randomised trials with a suspicion of bias (lower methodological quality) or meta-analyses of such trials or small bowel [12]. of trials with demonstrated heterogeneity • Familial GIST, i.e. families with a germ-line autoso- III Prospective cohort studies mal dominant mutation of KIT, are extremely rare, IV Retrospective cohort studies or case–control studies presenting with multiple GISTs at an early age. V Studies without control group, case reports, and experts’ opinions Grades of recommendation Diagnosis A Strong evidence for efficacy with a substantial clinical benefit, Clinical presentation and investigation strongly recommended The most common symptoms of GIST include upper gas - B Strong or moderate evidence for efficacy but with a limited clinical trointestinal bleeding and anaemia, whilst larger tumours benefit, generally recommended may present with abdominal pain/discomfort and a pal- C Insufficient evidence for efficacy or benefit does not outweigh the risk or the disadvantages (adverse events, costs,…), optional pable mass. Small bowel GISTs may remain silent for a D Moderate evidence against efficacy or for adverse outcome, generally long period before presenting with an acute event such not recommended as haemorrhage or rupture. Symptomatic colorectal E Strong evidence against efficacy or for adverse outcome, never GISTs may present with abdominal pain, obstruction recommended and lower gastrointestinal bleeding; oesophageal and gastro-oesophageal junction GISTs with dysphagia. Some patients may have non-specific systemic symptoms such as weight loss, night sweats and fever. Lack of awareness often multifocal, and with possible lymph node metasta- of the presenting features may lead to delayed diagnosis ses [5]. of GIST in some patients. Small GISTs may be asympto- matic and are often diagnosed incidentally during inves- Aetiology tigation for other conditions. In most cases the aetiology is unknown, although it is Small lesions below 2 cm in diameter may safely be fol- reported that patients with GIST are more likely to be lowed by endoscopic ultrasound on an annual basis and diagnosed with another cancer than the general popu- biopsied or excised if they continue to grow. For larger lation [6, 7], suggesting a likely link with inherited lesions it is necessary to make a histological diagnosis. The increased susceptibility to cancer in some patients. In standard approach for small nodules of 2  cm or above is the majority of cases GIST is associated with an acti- excisional biopsy, as GISTs of this size carry a higher risk. vating mutation in either the KIT or PDGFRA (platelet For gastric lesions it is preferable to perform a fine nee - derived growth factor receptor alpha) gene. However, dle aspirate or core needle biopsy under endoscopic ultra- other rare drivers may include mutations in NF1 (neu- sound (EUS) guidance. If this is not feasible, and limited rofibromatosis type 1, loss of function) or BRAF (gain surgery is possible, primary resection may be appropriate. of function). Tumours lacking mutations in KIT or However, the differential diagnosis of intra-abdominal PDGFRA are often called “wild-type”, and those lack- tumours may include leiomyosarcoma, germ cell tumour, ing mutations in not only these genes but also BRAF lymphoma, benign and malignant neurogenic tumours, and NF1 have been dubbed “quadruple wild-type” [8]. and fibromatosis. Given that the management of these Specific advice concerning the management of patients conditions differs substantially, and primary excision is with paediatric, “wild-type” and syndromic GIST can not always appropriate, it is sometimes necessary to do be obtained from a UK based alliance of medical spe- a percutaneous core needle biopsy to confirm the diag - cialists via the web site http://www.pawsgistclinic.org. nosis. This however carries a very small risk of contami - uk. nating the peritoneal cavity, especially if bleeding were to A number of genetic syndromes are linked to GIST: occur. If surgery would require multi visceral resection, or is likely to be morbid, e.g. total gastrectomy, then multiple • The Carney triad syndrome, comprising gastric GIST, core needle biopsies are definitely required, again ideally paraganglioma and pulmonary chondroma (these under EUS guidance, or alternatively using an ultrasound/ may occur at different ages) [9]. computed tomography (CT)-guided percutaneous • Carney–Stratakis syndrome, marked by germ-line approach. Depending on the histological diagnosis initial mutations of one of the SDH subunits A, B, C or D, Judson et al. Clin Sarcoma Res (2017) 7:6 Page 3 of 10 treatment may be with systemic therapy, e.g., lymphomas, sought, so that the tissue is available for later analyses and mesenteric fibromatosis, germ cell tumours and GIST, or research, provided that local ethical approval is in place. alternatively surveillance for benign entities such as non- Consent forms for the National GIST Tissue Bank based at malignant neurogenic tumours. Lesions at risk of rupture, the Royal Marsden Hospital can be found at http://www. such as cystic masses, should only be biopsied in special- givemysample.org/gist. ised centres. If a patient presents with obvious metastatic disease, then a biopsy of an easily accessible metastatic Key recommendations: focus should be performed and a laparotomy for diagnos- 1. Lesions larger than 2  cm in diameter need to be tic purposes is usually unnecessary. investigated and a diagnosis made. Often this can be Pathologically, the diagnosis of GIST relies on morpho- done by endoscopic ultrasound and needle biopsy, logical assessment and immunohistochemistry, the diag- particularly if the lesion is in the stomach. nosis being supported by CD117 immunopositivity [13, 2. The diagnosis should be made by a pathologist expe - 14]. More recently DOG1 has been added to the diagnostic rienced in the disease and include the use of immu- armamentarium [15, 16]. About 5% of GISTs are CD117 nohistochemistry and mutational analysis, which immunonegative, about 5% of GISTs are DOG1 immu- should be performed by an accredited laboratory. nonegative and about 1% of GISTs are immunonegative for 3. If neoadjuvant treatment with imatinib is planned, it both. The mitotic count has prognostic value and although is vital to confirm the diagnosis, since there is a wide several risk assessment tools use an index of mitoses per differential. It may be necessary to perform a percu - 50 high-power fields, it would be more accurate and repro - taneous core needle biopsy if the tumour is inacces- ducible to express this as the number of mitoses in a total sible to endoscopic ultrasound-guided biopsy. Muta- area of 5  mm , which is therefore recommended. If there tional analysis is obligatory, since some GISTs are is some diagnostic doubt, particularly in CD117 and/or insensitive to imatinib (e.g. those with D842V muta- DOG1 immunonegative suspected GIST, molecular analy- tion in exon 18 of PDGFRA). sis for activating mutations in KIT or PDGFRA may help confirm the diagnosis. Mutational analysis has predictive Risk assessment for primary tumours with no value for sensitivity to molecular-targeted therapy, and evidence of metastatic disease also prognostic value. If initial treatment is with imatinib, The TNM classification for staging has several limitations mutational analysis is particularly critical, since some and its use is not recommended in this disease. Prognos- GISTs are insensitive to the drug (e.g. PDGFRA exon 18 tic factors of proven value are the mitotic rate, tumour mutation D842V). The inclusion of mutational analysis in size, tumour site, and presence or absence of tumour the diagnostic work-up of all GISTs should be considered rupture. Gastric GISTs have a better prognosis than small standard practice, with the possible exclusion of sub 2 cm bowel or rectal GISTs. Tumour rupture through a sero- non-rectal GISTs, which are very unlikely ever to need sal surface is an adverse prognostic factor and should be medical treatment. It is strongly recommended that muta- recorded, whether it took place before or during surgery. tional analysis is performed in centralised laboratories Mutational status has not been incorporated in any risk which are enrolled in an external quality assurance pro- classification so far, although some genotypes have a dis - gram, and which have expertise in GIST analysis. In KIT/ tinct natural history, for example KIT/PDGFRA wild type PDGFRA mutation negative, or “Wild-Type” GIST, immu- GISTs tend to exhibit more indolent behaviour than KIT nohistochemistry for succinate dehydrogenase B (SDHB), exon 11 mutant disease. and, if negative, SDHA, should be performed if available, It is possible to assess risk of recurrence after resection since loss of expression may assist the diagnosis and may of a localised GIST. This can be especially useful in decid - help to guide therapy. In the absence of mutations in KIT ing the role of adjuvant therapy in individual patients. or PDGFRA it is also important to look for mutations in Several risk classifications have been proposed. An ini - BRAF, a rare, but important finding from a treatment tial risk classification developed at a consensus meeting perspective, since BRAF inhibitors are available, and also convened by the National Institute of Health (NIH) [13] in NF1. Patients with neurofibromatosis, who have a ger - was useful but only considered size and mitotic index. mline mutation in NF1, are at increased risk of GIST and Although the ‘high-risk’ category based on the NIH cri- this finding may indicate occult neurofibromatosis. The teria has a much worse prognosis than the others, with optimal treatment for advanced GIST with NF1 mutation both ‘very low-risk’ and ‘low-risk’ categories having a has yet to be determined. Collection of fresh/frozen tissue very favourable prognosis, the ‘intermediate-risk’ cat- is encouraged, because new molecular pathology assess- egory did not reliably identify patients with an unfavour- ments can then be made at a later stage in the patient’s able prognosis. A risk classification was proposed by the interest. Informed consent for tumour banking should be Armed Forces Institute of Pathology, which incorporates Judson et al. Clin Sarcoma Res (2017) 7:6 Page 4 of 10 Table 2 Modified NIH risk classification for primary GIST useful, mainly when early assessment of response to tyrosine kinase inhibitor therapy is of special interest, for Risk category Tumour size Mitotic index Primary tumour example after initiation of neo-adjuvant imatinib therapy. (cm) (per 50 HPFs) site Very low risk <2.0 ≤5 Any Treatment Low risk 2.1–5.0 ≤5 Any When small oesophago-gastric or duodenal nodules Intermediate 2.1–5.0 >5 Gastric less than 2  cm in size are detected, endoscopic biopsy risk <5.0 6–10 Any may be difficult and laparoscopic/laparotomic exci - 5.1–10.0 ≤5 Gastric sion may be the only way to make a histological diagno- High risk Any Any Tumour rupture sis. Many of these small nodules, if diagnosed as GIST, >10 Any Any will be low risk, or entities whose clinical significance >5.0 >5 Any remains unclear. Therefore, the standard approach to 2.1-5.0 >5 Non gastric these patients is endoscopic ultrasound assessment, usu- 5.1–10.0 ≤5 Non gastric ally with fine needle aspiration or core needle biopsy, After Joensuu [18] then annual follow-up, reserving excision for patients whose tumour increases in size or becomes symptomatic. Alternatively, the decision about a treatment plan can be the primary tumour site, in addition to mitotic count discussed with the patient. This may depend on age, life and tumour size, i.e. three of the main prognostic fac- expectancy and comorbidities. If follow-up is the choice, tors in localized GISTs [17]. There are also problems definitive evidence regarding an optimal surveillance pol - with the application of this risk classification tool since icy is lacking, but annual follow-up is reasonable. For a mitotic index and tumour size are non-linear continuous small histologically proven GIST of 2  cm or greater, the variables, so that the risk thresholds, especially based on standard treatment is excision, unless major morbidity number of mitoses, need to be interpreted with care. A is expected. Alternatively, in the case of a low-risk GIST, modification to the NIH criteria has been proposed that the possibility of surveillance could be discussed with the incorporates tumour site and additionally tumour rup- patient. Rectal (or recto-vaginal space) nodules should be ture, an important risk factor, [18] see Table 2. Prognostic biopsied and preferably excised after ultrasound assess- contour maps have been generated through a number of ment, regardless of tumour size. This is because GISTs at series of GIST patients not treated with adjuvant therapy. these sites have a higher risk of local recurrence after sur- These maps incorporate the mitotic index and tumour gery, and the local implications for surgery in relation to size as continuous non-linear variables, while tumour morbidity are more critical. In specific clinical contexts, rupture is considered in addition to tumour site. They if the tumour is small, a follow-up policy without surgery have been validated against pooled data from 10 series could be adopted, but such an approach should be dis- and 2560 patients from the literature [19]. Several nomo- cussed in detail with the patient. grams and web or mobile phone applications are avail- Multidisciplinary treatment planning is needed. This able to enable rapid risk category calculations to be made, should involve histopathologists, radiologists, surgeons, which may assist multidisciplinary planning of patient and oncologists, as well as gastroenterologists, nuclear management. medicine specialists and others, as applicable. Such teams are available in reference centres for sarcomas Staging procedures and GISTs, which treat a large number of GIST patients Staging procedures take into account the fact that most annually. Support staff, such as clinical nurse specialists, relapses affect the peritoneum and the liver. Contrast- play a vital role and are not likely to be available, or have enhanced abdominal and pelvic CT scan is the inves- the appropriate expertise, outside specialist centres. tigation of choice for staging and follow-up. Magnetic resonance imaging (MRI) or contrast-enhanced ultra- Localised disease—surgery sound may be alternatives, especially in younger patients Standard treatment of localized GISTs is complete surgi- where exposure to radiation should be limited. MRI pro- cal excision of the lesion, with no dissection of clinically vides better preoperative staging information for rectal negative lymph nodes [III, A]. Surgery should be per- GISTs. Chest CT scan or X-ray and routine laboratory formed by a sub speciality surgeon who is fully trained testing complement the staging work-up of the asymp- and experienced in radical anatomic site specific can - tomatic patient, but are not routinely required during cer surgery. When adjacent organs are involved, en bloc follow-up. Evaluation of F-fluorodeoxyglucose (FDG) resection is recommended wherever possible. If laparo- uptake using an FDG-positron emission tomography scopic excision is planned, the technique needs to follow (PET) scan, or FDG-PET–CT/MRI, can sometimes be Judson et al. Clin Sarcoma Res (2017) 7:6 Page 5 of 10 the principles of oncological surgery [20] [III, A]. A lap- and was approved by the National Institute for Care and aroscopic approach is clearly discouraged in patients Health Excellence (NICE) in their recent re-appraisal with large tumours, because of the risk of tumour rup- (https://www.nice.org.uk/guidance/ta326). Mutational ture, which is associated with a very high risk of relapse. analysis is critical to making a clinical decision regard- R0 excision is the goal (i.e., an excision whose margins ing adjuvant therapy. In fact, there is a consensus that are clear of tumour cells). When R0 surgery is likely to PDGFRA D842  V-mutated GISTs should not be treated result in major functional sequelae, e.g. total gastrec- with any adjuvant therapy, given the lack of sensitivity of tomy or abdomino-perineal resection of rectum, neo- this genotype to imatinib both in  vitro and in  vivo [IV, adjuvant imatinib should be regarded as standard therapy A]. Given the data supporting the use of a higher dose of [21–24] [IV, A]. Treatment is given to reduce the size imatinib (800 mg daily) in the presence of an exon 9 KIT of the tumour and to limit subsequent surgical mor- mutation in advanced GIST, clinicians might consider bidity. This may also be the case if surgery will be safer using this dose in the adjuvant setting for this genotype following cyto-reduction (e.g. the risk of bleeding and [28–31]. However, this is not supported by any controlled tumour rupture is likely to be decreased). After maxi- trial data in the adjuvant setting and use of the higher mal tumour response, generally after 6–12  months, sur- dose is not approved by NICE in the UK. There is con - gery is performed. Prior mutational analysis is crucial to sensus on avoiding adjuvant treatment in NF-1 related prevent patients with less sensitive or resistant tumours GISTs, which are insensitive to imatinib in the advanced (e.g. PDGFRA D842  V mutations) from receiving ther- setting. On the other hand, a consensus is lacking among apy with imatinib, and to allow appropriate dosing for experts about whether KIT/PDGFRA wild-type SDH- patients with KIT exon 9 mutated tumours. Early tumour negative GIST should be treated with adjuvant therapy. response assessment is mandatory, so that surgery is not This reflects their lower sensitivity to imatinib, as well as delayed in the case of non-responding disease. Functional their peculiar natural history, which is often more indo- imaging, such as PET-CT, makes it possible to assess the lent; subgroup analyses of available randomized trials are tumour response very rapidly, within a few weeks. There too limited to provide sufficient evidence. are limited data to guide the physician on when to stop If there has been tumour rupture before or during sur- imatinib before surgery, but it can safely be stopped a few gery, there will have been spillage of tumour cells into the days, or even 1 day, before surgery and it can be resumed peritoneal cavity, and therefore occult peritoneal disease promptly when the patient has recovered from the acute can be assumed to exist. This puts the patient at a very effects of surgery. high risk of peritoneal relapse. Therefore, these patients If preoperative medical treatment has not helped or should be considered for adjuvant imatinib therapy. The cannot be used, there should be a discussion with the optimal duration of treatment in these cases is unknown, patient about accepting a possible R1 resection with given the uncertainty as to whether they should be microscopically positive margins (i.e. excision mar- viewed as essentially having metastatic disease, but gins containing tumour cells) [IV, B]. This may be more should be at least 3 years, as for high risk resected GIST. acceptable for low-risk lesions, with the lack of any for- mal demonstration that R1 surgery is associated with Key recommendations worse overall survival [25]. If an unplanned R1 exci- 1. GIST should be managed by an experienced multi- sion has already been carried out, re-excision may be an disciplinary team. option, provided the original site of lesion can be found, 2. Pre-operative imatinib should be considered for and major functional sequelae are not foreseen. those large gastric or rectal primaries where immedi- ate resection is likely to be morbid, e.g. total gastrec- Localised disease—adjuvant therapy tomy or abdomino-perineal resection. In this situ- The risk of relapse following surgery can be substan - ation mutational analysis is mandatory prior to the tial, as defined by available risk classifications. Adjuvant initiation of imatinib therapy. treatment with imatinib for 3  years was associated with 3. Patients at high risk of recurrence or distant relapse improved relapse-free and overall survival compared should receive 3 years of adjuvant imatinib, provided with 1  year of therapy in a randomized trial in high-risk their tumour is not likely to be resistant to therapy patients [26]. Previously, a placebo-controlled trial dem- (PDGFRA exon 18 mutation D842V). onstrated that imatinib given for 1 year prolongs relapse- free survival in localized GISTs larger than 3  cm with a Metastatic disease—systemic treatment macroscopically complete resection [27]. Therefore, adju - In patients with inoperable and metastatic disease, vant therapy with imatinib for 3  years is standard treat- imatinib is the standard treatment [32, 33] [III, A], includ- ment for patients with a significant risk of relapse [I, A] ing patients who had previously received the drug as Judson et al. Clin Sarcoma Res (2017) 7:6 Page 6 of 10 adjuvant therapy without relapse during this treatment. shown to be related to a good prognosis, provided the This also applies to metastatic patients whose disease has patient is responding to imatinib, but whether this is due been completely removed surgically, although surgery to surgery or to patient selection [38–40] has never been as a primary approach to metastatic GIST is not recom- demonstrated prospectively. Conducting a randomised mended. The standard dose of imatinib is 400  mg daily trial did not prove feasible; thus, at the present time sur- [I, A]. However, data have shown that patients with KIT gery can be discussed with the patient but not recom- exon 9 mutations fare better in terms of progression-free mended on the basis of a definitive proof of benefit [III, survival (PFS) on a higher dose of 800 mg daily, which is C]. Surgical excision of progressive disease is not recom- therefore the standard treatment in this subgroup [31] mended, given the poor results in published series, but [III, A], albeit not recommended by a NICE appraisal surgery of limited progression, such as the ‘nodule within which only assessed dose escalation in the context of dis- a mass’, has been associated with a progression-free inter- ease progression. A report on the long term follow-up of val in the same range as for second-line treatment with the European/Australasian clinical trial which compared sunitinib. So, this may be a palliative option in the indi- 400  mg with 800  mg imatinib in patients with advanced vidual patient with limited progression, while continuing GIST has shown a survival advantage for the initial use imatinib [V, C]. Non-surgical procedures, such as radi- of the 800 mg dose in those with exon 9 mutations in KIT ofrequency ablation of liver metastases may also be used. (Casali P et al., in press) indicating a need to review this Prior to performing such interventions PET-CT can issue in the UK. Treatment should be continued indefi - be useful to confirm the location of imatinib-resistant nitely, since treatment interruption is generally followed disease. by relatively rapid tumour progression, even when lesions Dose escalation of imatinib to 800  mg in the case of a have previously been surgically excised [34] [II, B]. At GIST with a KIT exon 9 mutation showing disease pro- the start of treatment the patient should be alerted to gression could be considered if the higher dose was not the importance of adherence to therapy, and of possible used initially, since the higher dose is significantly more interactions with concomitant medications and foods. effective in this setting [31]. Higher doses, though not They should also be given guidance about the best ways necessarily 800 mg, could be useful if satisfactory plasma to handle any possible side effects. Dose intensity should levels of imatinib are not being achieved, but the use of be maintained by effective management of side effects, higher doses is not approved by NICE. The potential and a rational policy of dose reductions and interruptions misinterpretation of the images produced by the com- should be applied if there is excessive, persistent toxicity. plex tumour response patterns to TKIs can lead to a false Retrospective data suggest that suboptimal plasma levels diagnosis of progression, which must be considered. of imatinib are associated with a worse outcome, though Patient non-compliance and drug interactions with con- a correlation with outcome has not been established pro- comitant medications must also be ruled out as the pos- spectively [35]. A recent report confirmed that patients sible cause of progression. with imatinib trough levels of less than 760 ng/ml, taken If there is confirmed progression, or rare intolerance after a minimum of 3  months’ treatment, which equates to imatinib after all attempts to manage side effects have to steady state [36], had a worse outlook in terms of pro- failed, the standard second-line treatment is the tyros- gression-free survival, which applied in the case of both ine kinase inhibitor (TKI) sunitinib [41] [I, B]. This drug gastric and small bowel GIST [37]. Aside from its poten- was proven to be effective in terms of PFS using a regi - tial use to tailor the imatinib dose, plasma level assess- men of 50  mg daily 4  weeks on/2  weeks off. Data have ment may be useful in the case of: (i) patients receiving been published showing that continuous treatment with concomitant medications that put them at a risk of major a lower daily dose of 37.5  mg is also effective and well interactions; (ii) unexpected observed toxicities; (iii) tolerated, although no formal comparison has been per- progression on 400  mg. Dose adaptation according to formed within a randomized clinical trial. This schedule inadequate imatinib trough level is being studied in the can therefore be considered an alternative on an individ- Netherlands, and is a standard approach in a number of ualized basis [42] [III, B]. However, not all patients resist- institutions. However, the use of a higher dose of imatinib ant to imatinib respond to sunitinib particularly those in patients with progressive disease is not approved by with secondary mutations affecting the activation loop NICE. domain of KIT and the PDGFRA exon 18 D842V muta- Close monitoring of the tumour response should be tion, which is always resistant. carried out in the early phases of treatment. Follow-up A prospective placebo-controlled randomized trial should be continued throughout the treatment, since the demonstrated that regorafenib, at a dose of 160 mg daily risk of secondary progression persists over time. Com- on a 3  weeks on/1  week off schedule, significantly pro - plete excision of residual metastatic disease has been longed PFS in patients progressing after both imatinib Judson et al. Clin Sarcoma Res (2017) 7:6 Page 7 of 10 and sunitinib [43]. Regorafenib is regarded as standard Response assessment therapy for the third-line treatment of patients progress- Response assessment is complex and early progression ing on or failing to respond to imatinib and sunitinib [I, in particular should be confirmed by a team experienced B]. It is currently available in England via the National in treating GIST. Anti-tumour activity translates into Cancer Drugs Fund and is also available in Scotland, tumour shrinkage in most patients, but some patients Wales and Northern Ireland as standard 3rd line therapy. may show only changes in tumour “density” on imaging, The key distinction between sunitinib and regorafenib, these changes sometimes precede a reduction in tumour as also previously shown with the analogue sorafenib, is volume. Such changes in tumour radiological appearance its ability to inhibit tumours with secondary mutations in should be considered as indicative of tumour response. the activation loop of KIT, especially in exon 17 [44, 45]. Tumour size may even increase in the short term but if These mutations are known to confer resistance both to tumour density on CT scan is decreased this may still imatinib and sunitinib, hence the value of regorafenib in indicate tumour response [48, 49]. Even the apparent this setting. ‘appearance’ of new lesions may be due to them becom- Patients with metastatic GIST failing all three standard ing less dense, or cystic, especially in the liver. Therefore, agents should be considered for participation in clinical both tumour size and tumour density on CT scan, or trials of new agents. These studies are only likely to be consistent changes on MRI or contrast-enhanced ultra- available in major centres treating GIST. There is limited sound, should be considered when determining tumour evidence that patients who have already progressed on response. F-FDG-PET has proved useful in the early imatinib may benefit for a limited period when re-chal - assessment of tumour response, for example when pre- lenged with the drug [46]. Likewise, there is anecdotal diction of the response is valuable, for example in the case evidence that maintaining treatment with a TKI even in of preoperative therapy, or when response is in doubt. the case of progressive disease, as opposed to stopping it, However, a small proportion of GISTs have no FDG may slow down progression if no other option is available uptake. The absence of tumour progression at 6  months at the time. Therefore, re-challenging or continuing treat - [50] is also equivalent to a tumour response. Conversely, ment with a TKI, to which the patient has already been tumour progression may not always be accompanied by exposed, is an option which may be considered for symp- changes in tumour size. For example, an increase in the tom control in patients with progression [V, B]. tumour density shown by contrast enhancement within a previously responding low density tumour lesion, may be Metastatic disease—local therapy indicative of tumour progression. A typical progression Selected patients with limited liver metastatic disease pattern is the ‘nodule within the mass’, in which a portion may be amenable to surgery or radiofrequency ablation of a responding lesion becomes hyper-dense [51]. (RFA) after maximum response to imatinib, or if there is evidence of localised disease progression. The use of RFA Key recommendations is restricted to tumours in the region of 3  cm in maxi- 1. Imatinib is the treatment of choice for patients with mum diameter and is less likely to be a suitable approach unresectable or metastatic disease and is given until for lesions adjacent to large vessels or superficial lesions, progression at the standard dose of 400  mg daily. especially if displacing the liver capsule. However, larger Data suggest that patients whose tumours have an isolated lesions and superficial lesions may still be suit - exon 9 mutation in KIT benefit from a larger dose, able for surgical resection, either by partial hepatectomy though this is not currently recommended by NICE. or wedge resection. Dedicated liver MRI scans and when 2. Isolated progression may be amenable to surgery or appropriate PET-CT scans may be required to determine other local measures, such as radiofrequency abla- whether this is a legitimate approach by excluding other tion. occult active disease. 3. Standard second line treatment is sunitinib, which Radiotherapy can be a useful local therapy in GIST may be given at the recommended dose of 50  mg under certain circumstances in the advanced disease set- daily for 4 weeks every 6 weeks, or 37.5 mg daily con- ting. If there is a single site of disease that is progressing tinuously. on a TKI and can be encompassed within a radiotherapy 4. Standard 3rd line treatment is regorafenib. treatment field, then radiotherapy delivered to a moder - ate or high dose can offer local tumour control, and pos - Follow‑up sibly prolong the use of the TKI [47]. Radiotherapy can The optimal follow-up policy for surgically treated also be used at lower doses to palliate symptomatic dis- patients with localized disease is unclear. Relapses occur ease, for example to relieve pain or bleeding. most often in the liver and/or the peritoneal cavity. Other sites of metastases, including bone and brain are Judson et al. Clin Sarcoma Res (2017) 7:6 Page 8 of 10 Authors’ contributions uncommon, but may be less unusual following prolonged All authors contributed to the content of the manuscript and consent to its treatment with several lines of therapy. The mitotic rate publication. IJ adapted the guideline from the previous version published in most likely affects the frequency with which relapses 2010, with reference to the ESMO Guidelines from 2014 of which he was a co-author. Amendments and suggestions from the co-authors were incorpo- occur. Risk assessment based on the mitotic count, rated. All authors read and approved the final manuscript. tumour size and tumour site may be useful in choosing the routine follow-up policy. High-risk patients generally Author details The Institute of Cancer Research, Royal Marsden NHS Foundation Trust, relapse within 1–3  years from the end of adjuvant ther- Fulham Road, London SW3 6JJ, UK. Addenbrooke’s Hospital, Cambridge apy. Low-risk patients may relapse later, given that the University Hospitals, Cambridge, UK. University College Hospital, University disease is likely to be slower growing. College London Hospitals NHS Foundation Trust, London, UK. Bristol Cancer Institute, University Hospitals, Bristol NHS Trust, Bristol, UK. Southmead The issue of follow-up has been addressed by Joen - Hospital, North Bristol NHS Trust, Bristol, UK. Royal Marsden NHS Foundation suu and colleagues based on the currently available Trust London, London, UK. evidence [52]. High-risk patients who have undergone Acknowledgements resection of their primary generally undergo a rou- The authors would also like to thank the following on behalf of the British tine follow-up with abdominal CT or MRI scan every Sarcoma Group for valuable review, suggestions and amendments: Dr. Robin 3–6  months during adjuvant therapy, for 3  years. This Jones (Consultant Medical Oncologist, Royal Marsden Hospital, London), Dr. Palma Dileo (Consultant Medical Oncologist, University College Hospital, frequency of follow up is because of the need to man- London), Mrs. Judith Robinson (GIST Support UK), Jayne Bressington (GIST age the side effects of the therapy. On cessation of adju - Support UK), Dr. Nicholas Carroll (Consultant Radiologist, Addenbrooke’s vant therapy follow up is every 3  months for 2  years, Hospital, Cambridge). then every 6  months for another 3  years, after which Competing interests follow up is annually for another 5 years. Patients with IJ has received honoraria for attending advisory boards and speaking engage- high-risk tumours not given adjuvant therapy, for ments from GSK, Ariad, Amgen, Clinigen, Lilly and Bayer; BS has received honoraria for attending advisory boards from Ariad, Novartis, Clinigen, Lilly whatever reason, should be followed up 3 monthly for and Daichi. NW has received honoraria from Novartis for speaking engage- 2  years, 6 monthly for 3  years and then annually for a ments; RB has received honoraria for attending advisory boards and speaking further 5 years. engagements from Novartis, Pfizer, Ariad, Pierre Fabre, Bristol Myers Squibb, Astra Zeneca, and Bayer Pharmaceuticals. AD & SM have no competing For low to intermediate risk tumours, the optimal interests to report. frequency of follow-up is less clear. If follow-up is per- formed, it will usually be an abdominal CT or MRI scan, Availability of data No clinical or biological data was included in the preparation of these or ultrasound, every 6–12 months for 5 years. guidelines. Very low-risk GISTs do not require routine follow-up, provided excision was complete, although one must be Ethics approval and consent to participate These guidelines do not contain information on patients, hence neither ethi- aware that the risk of progression is not zero. cal approval nor consent from patients was required. Radiation exposure is a factor to consider when select- ing the imaging modality for long term follow-up. Publisher’s Note Abdominal MRI is an acceptable alternative to CT which Springer Nature remains neutral with regard to jurisdictional claims in pub- could be used at selected intervals. lished maps and institutional affiliations. Received: 2 February 2017 Accepted: 6 April 2017 Key recommendations 1. Patients with high risk disease on adjuvant therapy should be followed up by cross-sectional imaging every 3–6 months during their 3 years of treatment, 3 monthly for 2  years following cessation of treat- References 1. Group ESESNW. Gastrointestinal stromal tumours: ESMO clinical practice ment and thereafter every 6  months for 3  years and guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2014;25 annually for 5 years. Suppl 3:iii21–6. 2. Patients with high risk disease not receiving adjuvant 2. Dykewicz CA, Centers for Disease C, Prevention, Infectious Diseases Society of A, American Society of B, Marrow T. Summary of the guidelines treatment should follow the post-adjuvant treatment for preventing opportunistic infections among hematopoietic stem cell scheme. transplant recipients. Clin Infect Dis. 2001;33(2):139–44. 3. For patients with lesser levels of risk less frequent fol- 3. Gatta G, van der Zwan JM, Casali PG, Siesling S, Dei Tos AP, Kunkler I, et al. Rare cancers are not so rare: the rare cancer burden in Europe. Eur J low-up is generally recommended, although the clini- Cancer. 2011;47(17):2493–511. cal benefits are unclear. For intermediate risk patients 4. Cassier PA, Ducimetiere F, Lurkin A, Ranchere-Vince D, Scoazec JY, Bringuier 6 monthly scans for 5 years followed by annual scans PP, et al. A prospective epidemiological study of new incident GISTs during two consecutive years in Rhone Alpes region: incidence and molecular and for low risk patients shorter duration follow-up distribution of GIST in a European region. Br J Cancer. 2010;103(2):165–70. would be considered reasonable. 5. Pappo AS, Janeway KA. Pediatric gastrointestinal stromal tumors. Hema- tol Oncol Clin N Am. 2009;23(1):15–34. Judson et al. Clin Sarcoma Res (2017) 7:6 Page 9 of 10 6. Agaimy A, Wunsch PH, Sobin LH, Lasota J, Miettinen M. Occurrence of 27. Dematteo RP, Ballman KV, Antonescu CR, Maki RG, Pisters PW, Demetri other malignancies in patients with gastrointestinal stromal tumors. GD, et al. Adjuvant imatinib mesylate after resection of localised, primary Semin Diagn Pathol. 2006;23(2):120–9. gastrointestinal stromal tumour: a randomised, double-blind, placebo- 7. Ballinger ML, Goode DL, Ray-Coquard I, James PA, Mitchell G, Niedermayr controlled trial. Lancet. 2009;373(9669):1097–104. E, et al. Monogenic and polygenic determinants of sarcoma risk: an 28. Debiec-Rychter M, Sciot R, Le Cesne A, Schlemmer M, Hohenberger P, international genetic study. Lancet Oncol. 2016;17:1261–71. van Oosterom AT, et al. KIT mutations and dose selection for imatinib in 8. Pantaleo MA, Nannini M, Corless CL, Heinrich MC. Quadruple wild-type patients with advanced gastrointestinal stromal tumours. Eur J Cancer. ( WT ) GIST: defining the subset of GIST that lacks abnormalities of KIT, 2006;42(8):1093–103. PDGFRA, SDH, or RAS signaling pathways. Cancer Med. 2015;4(1):101–3. 29. Heinrich MC, Owzar K, Corless CL, Hollis D, Borden EC, Fletcher CD, 9. Zhang L, Smyrk TC, Young WF Jr, Stratakis CA, Carney JA. Gastric stromal et al. Correlation of kinase genotype and clinical outcome in the North tumors in Carney triad are different clinically, pathologically, and behavio - American Intergroup Phase III Trial of imatinib mesylate for treatment of rally from sporadic gastric gastrointestinal stromal tumors: findings in 104 advanced gastrointestinal stromal tumor: CALGB 150105 Study by Cancer cases. Am J Surg Pathol. 2010;34(1):53–64. and Leukemia Group B and Southwest Oncology Group. J Clin Oncol. 10. Pasini B, McWhinney SR, Bei T, Matyakhina L, Stergiopoulos S, Muchow 2008;26(33):5360–7. M, et al. Clinical and molecular genetics of patients with the Carney– 30. Heinrich MC, Corless CL, Demetri GD, Blanke CD, von Mehren M, Joensuu Stratakis syndrome and germline mutations of the genes coding for the H, et al. Kinase mutations and imatinib response in patients with meta- succinate dehydrogenase subunits SDHB, SDHC, and SDHD. Eur J Hum static gastrointestinal stromal tumor. J Clin Oncol. 2003;21(23):4342–9. Genet. 2008;16(1):79–88. 31. Gastrointestinal Stromal Tumor Meta-Analysis G. Comparison of two 11. Gaal J, Stratakis CA, Carney JA, Ball ER, Korpershoek E, Lodish MB, et al. doses of imatinib for the treatment of unresectable or metastatic gastro- SDHB immunohistochemistry: a useful tool in the diagnosis of Carney– intestinal stromal tumors: a meta-analysis of 1640 patients. J Clin Oncol. Stratakis and Carney triad gastrointestinal stromal tumors. Mod Pathol. 2010;28(7):1247–53. 2011;24(1):147–51. 32. Verweij J, Casali PG, Zalcberg J, LeCesne A, Reichardt P, Blay JY, et al. 12. Miettinen M, Fetsch JF, Sobin LH, Lasota J. Gastrointestinal stromal tumors Progression-free survival in gastrointestinal stromal tumours with high- in patients with neurofibromatosis 1: a clinicopathologic and molecular dose imatinib: randomised trial. Lancet. 2004;364(9440):1127–34. genetic study of 45 cases. Am J Surg Pathol. 2006;30(1):90–6. 33. Blanke CD, Rankin C, Demetri GD, Ryan CW, von Mehren M, Benjamin RS, 13. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, et al. et al. Phase III randomized, intergroup trial assessing imatinib mesylate at Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum two dose levels in patients with unresectable or metastatic gastrointes- Pathol. 2002;33(5):459–65. tinal stromal tumors expressing the kit receptor tyrosine kinase: S0033. J 14. Rubin BP, Cooper K, Fletcher CD, Folpe AL, Gannon FH, Hunt JL, et al. Clin Oncol. 2008;26(4):626–32. Protocol for the examination of specimens from patients with tumors of 34. Le Cesne A, Ray-Coquard I, Bui BN, Adenis A, Rios M, Bertucci F, et al. soft tissue. Arch Pathol Lab Med. 2010;134(4):e31–9. Discontinuation of imatinib in patients with advanced gastrointestinal 15. Miettinen M, Wang ZF, Lasota J. DOG1 antibody in the differential diag- stromal tumours after 3 years of treatment: an open-label multicentre nosis of gastrointestinal stromal tumors: a study of 1840 cases. Am J Surg randomised phase 3 trial. Lancet Oncol. 2010;11(10):942–9. Pathol. 2009;33(9):1401–8. 35. Demetri GD, Wang Y, Wehrle E, Racine A, Nikolova Z, Blanke CD, et al. 16. Lopes LF, West RB, Bacchi LM, van de Rijn M, Bacchi CE. DOG1 for Imatinib plasma levels are correlated with clinical benefit in patients with the diagnosis of gastrointestinal stromal tumor (GIST ): comparison unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol. between 2 different antibodies. Appl Immunohistochem Mol Morphol. 2009;27(19):3141–7. 2010;18(4):333–7. 36. Eechoute KFM, Reyners AK, de Jong FA, Sparreboom A, van der Graaf 17. Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and WTA, et al. A long-term prospective population pharmacokinetic study prognosis at different sites. Semin Diagn Pathol. 2006;23(2):70–83. on imatinib plasma concentrations in GIST patients. Clin Cancer Res. 18. Joensuu H. Risk stratification of patients diagnosed with gastrointestinal 2012;18:5780–7. stromal tumor. Hum Pathol. 2008;39(10):1411–9. 37. Bouchet S, Poulette S, Titier K, Moore N, Lassalle R, Abouelfath A, et al. 19. Joensuu H, Vehtari A, Riihimaki J, Nishida T, Steigen SE, Brabec P, et al. Risk Relationship between imatinib trough concentration and outcomes in of recurrence of gastrointestinal stromal tumour after surgery: an analysis the treatment of advanced gastrointestinal stromal tumours in a real-life of pooled population-based cohorts. Lancet Oncol. 2012;13(3):265–74. setting. Eur J Cancer. 2016;57:31–8. 20. Novitsky YW, Kercher KW, Sing RF, Heniford BT. Long-term outcomes of 38. Raut CP, Posner M, Desai J, Morgan JA, George S, Zahrieh D, et al. Surgical laparoscopic resection of gastric gastrointestinal stromal tumors. Ann management of advanced gastrointestinal stromal tumors after treat- Surg. 2006;243(6):738–45 (discussion 45–7). ment with targeted systemic therapy using kinase inhibitors. J Clin Oncol. 21. Eisenberg BL, Harris J, Blanke CD, Demetri GD, Heinrich MC, Watson JC, 2006;24(15):2325–31. et al. Phase II trial of neoadjuvant/adjuvant imatinib mesylate (IM) for 39. Gronchi A, Fiore M, Miselli F, Lagonigro MS, Coco P, Messina A, et al. advanced primary and metastatic/recurrent operable gastrointestinal Surgery of residual disease following molecular-targeted therapy stromal tumor (GIST ): early results of RTOG 0132/ACRIN 6665. J Surg with imatinib mesylate in advanced/metastatic GIST. Ann Surg. Oncol. 2009;99(1):42–7. 2007;245(3):341–6. 22. Cananzi FC, Judson I, Lorenzi B, Benson C, Mudan S. Multidisciplinary care 40. Mussi C, Ronellenfitsch U, Jakob J, Tamborini E, Reichardt P, Casali PG, of gastrointestinal stromal tumour: a review and a proposal for a pre- et al. Post-imatinib surgery in advanced/metastatic GIST: is it worthwhile treatment classification. Eur J Surg Oncol. 2013;39(11):1171–8. in all patients? Ann Oncol. 2010;21(2):403–8. 23. Wilkinson MJ, Fitzgerald JE, Strauss DC, Hayes AJ, Thomas JM, Messiou C, 41. Demetri GD, van Oosterom AT, Garrett CR, Blackstein ME, Shah MH, et al. Surgical treatment of gastrointestinal stromal tumour of the rectum Verweij J, et al. Efficacy and safety of sunitinib in patients with advanced in the era of imatinib. Br J Surg. 2015;102(8):965–71. gastrointestinal stromal tumour after failure of imatinib: a randomised 24. Rutkowski P, Gronchi A, Hohenberger P, Bonvalot S, Schoffski P, Bauer controlled trial. Lancet. 2006;368(9544):1329–38. S, et al. Neoadjuvant imatinib in locally advanced gastrointestinal 42. George S, Blay JY, Casali PG, Le Cesne A, Stephenson P, Deprimo SE, et al. stromal tumors (GIST ): the EORTC STBSG experience. Ann Surg Oncol. Clinical evaluation of continuous daily dosing of sunitinib malate in 2013;20(9):2937–43. patients with advanced gastrointestinal stromal tumour after imatinib 25. McCarter MD, Antonescu CR, Ballman KV, Maki RG, Pisters PW, Demetri failure. Eur J Cancer. 2009;45(11):1959–68. GD, et al. Microscopically positive margins for primary gastrointestinal 43. Demetri GD, Reichardt P, Kang YK, Blay JY, Rutkowski P, Gelderblom H, stromal tumors: analysis of risk factors and tumor recurrence. J Am Coll et al. Efficacy and safety of regorafenib for advanced gastrointestinal Surg. 2012;215(1):53–9 (discussion 9–60). stromal tumours after failure of imatinib and sunitinib (GRID): an interna- 26. Joensuu H, Eriksson M, Sundby Hall K, Hartmann JT, Pink D, Schutte J, tional, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet. et al. One vs three years of adjuvant imatinib for operable gastrointestinal 2013;381(9863):295–302. stromal tumor: a randomized trial. JAMA. 2012;307(12):1265–72. 44. Heinrich MC, Marino-Enriquez A, Presnell A, Donsky RS, Griffith DJ, McKinley A, et al. Sorafenib inhibits many kinase mutations associated Judson et al. Clin Sarcoma Res (2017) 7:6 Page 10 of 10 with drug-resistant gastrointestinal stromal tumors. Mol Cancer Ther. tomography in patients with metastatic gastrointestinal stromal 2012;11(8):1770–80. tumor treated at a single institution with imatinib mesylate: pro- 45. Hsueh YS, Lin CL, Chiang NJ, Yen CC, Li CF, Shan YS, et al. Selecting tyros- posal of new computed tomography response criteria. J Clin Oncol. ine kinase inhibitors for gastrointestinal stromal tumor with secondary 2007;25(13):1753–9. KIT activation-loop domain mutations. PLoS ONE. 2013;8(6):e65762. 50. Le Cesne A, Van Glabbeke M, Verweij J, Casali PG, Findlay M, Reichardt P, 46. Kang YK, Ryu MH, Yoo C, Ryoo BY, Kim HJ, Lee JJ, et al. Resumption of et al. Absence of progression as assessed by response evaluation criteria imatinib to control metastatic or unresectable gastrointestinal stromal in solid tumors predicts survival in advanced GI stromal tumors treated tumours after failure of imatinib and sunitinib (RIGHT ): a randomised, with imatinib mesylate: the intergroup EORTC-ISG-AGITG phase III trial. J placebo-controlled, phase 3 trial. Lancet Oncol. 2013;14(12):1175–82. Clin Oncol. 2009;27(24):3969–74. 47. Joensuu H, Eriksson M, Collan J, Balk MH, Leyvraz S, Montemurro M. 51. Shankar S, vanSonnenberg E, Desai J, Dipiro PJ, Van Den Abbeele A, Radiotherapy for GIST progressing during or after tyrosine kinase inhibi- Demetri GD. Gastrointestinal stromal tumor: new nodule-within-a-mass tor therapy: a prospective study. Radiother Oncol. 2015;116(2):233–8. pattern of recurrence after partial response to imatinib mesylate. Radiol- 48. Benjamin RS, Choi H, Macapinlac HA, Burgess MA, Patel SR, Chen ogy. 2005;235(3):892–8. LL, et al. We should desist using RECIST, at least in GIST. J Clin Oncol. 52. Joensuu H, Martin-Broto J, Nishida T, Reichardt P, Schoffski P, Maki RG. 2007;25(13):1760–4. Follow-up strategies for patients with gastrointestinal stromal tumour 49. Choi H, Charnsangavej C, Faria SC, Macapinlac HA, Burgess MA, Patel treated with or without adjuvant imatinib after surgery. Eur J Cancer. SR, et al. Correlation of computed tomography and positron emission 2015;51(12):1611–7. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit

Journal

Clinical Sarcoma Research – Springer Journals

Published: Apr 21, 2017

There are no references for this article.