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

Learn More →

The TGF-β1 dynamics during radiation therapy and its correlation to symptomatic radiation pneumonitis in lung cancer patients

The TGF-β1 dynamics during radiation therapy and its correlation to symptomatic radiation... Backgroud: The underlying molecular and cellular mechanisms of radiation pneumonitis (RP) are very complex. Several biological factors need to be considered together with the well known dosimetric parameters for understanding the molecular events in developing RP in lung cancer patients. The aim of this study was to correlate the variations of the cytokine levels in lung cancer patients during radiation therapy (RT) with the occurrence of symptomatic RP. Methods: Thirty-four lung cancer patients who received three-dimensional conformal radiation therapy were evaluated prospectively. Serial blood samples before, at the beginning, in the middle of, at the end of RT and 2 and 4 weeks after RT were analyzed for IL-1α, IL-6, IL-10, TNF-α and TGF-β1 by performing enzyme-linked immunosorbent assay. The predictive values of dosimetric factors for RP were evaluated, too. Results: Overall, 8 patients (23.5%) had grade ≥ 2 RP. By serial measurement of cytokines level, only the TGF-β1 level showed a correlation to the symptomatic RP. None of the other cytokines, IL-1α, IL-6, IL-10 and TNF-α level was correlated with the risk of RP. The mean pretreatment TGF- β1 level did not differ between RP and non-RP groups. However, during the period of radiation treatment, the TGF-β1 level began to increase at the end of RT in the RP group and became significantly higher 4 weeks after RT (p = 0.007). Using an ANOVA model for repeated-measures, we found significant associations between the changes of TGF-β1 during the time course of the RT and the risk of developing RP (p < 0.001). Most of the dosimetric factors showed a significant association with RP. Conclusion: Our results show that the changes of TGF-β1 could be correlated with RP and the incorporation of the biological parameters into the dosimetric data could be useful for predicting symptomatic RP. Page 1 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 A chest CT scan including the entire lung volume was per- Background Increasing radiation dose and combining it with chemo- formed with the immobilization device at 3 mm scan therapy were found to improve the local control for lung thickness. The structures of interest, such as gross tumor cancer and thus improve the overall survival of lung can- volume (GTV), clinical target volume (CTV), and normal cer patients [1-4]. However, the tolerance of the surround- structures (esophagus, spinal cord, heart, and lung) were ing normal tissues to radiation therapy limits the level of defined and contoured on the multiple CT images. The dose that can be delivered for the treatment of lung can- gross tumor volume included the tumor and clinically cer. involved nodes. GTV was determined by chest CT and/or PET-CT. CTV included GTV with additional uniform 5 cm The risk of radiation pulmonary toxicity may increase for expansion and ipsilateral hilar lymph nodes. The unin- the following patients: patients with a poor performance volved mediastinal, and supraclavicular nodal regions status or inadequate pulmonary function, patients who were not routinely included in the CTV, but high risk undergo combined chemoradiotherapy and patients nodal stations near GTV nodes were included occasion- receiving an increased total radiation dose and treatment ally. The PTV was obtained by a 0.7~1.0 cm margin 3-D volume. Dosimetric parameters such as V5, V10, V20, expansion from the CTV. An additional 0.5~1.0 cm mar- V30, V40, the mean lung dose (MLD) and the normal tis- gin was added to superior and inferior direction to sue complications probability (NTCP) have been reported account for respiratory motion as assessed under fluoros- to be related with RP by several retrospective or prospec- copy. The beam arrangement was planned to minimize tive studies [5-11]. However, the underlying molecular the irradiated lung volume usually using 3 to 5 coplanar and cellular mechanisms of RP are very complex. Several oblique beams. Some patients of having bulky tumor biological factors need to be considered together with the were treated with AP/PA fields to include the CTV for the above mentioned dosimetric parameters for understand- first 30~40 Gy, followed by off-cord oblique beam to the ing the molecular events in developing radiation-induced GTV usually composed of 3~5 beams. After making cor- complications in normal tissue [5,8,12]. Several clinical rection for the tissue inhomogeneity using convolution/ reports have suggested a possible role of profibrogenic superposition algorithm of the treatment planning system and proinflammatory cytokines in the modulation of (Pinnacle, Philips Medical System, Andover, MA, U.S.A.), radiation pulmonary injury [12-18]. the mean lung dose (MLD), V5, V10, V20 and V30 (the percentages of the irradiated lung volume receiving a radi- We designed a prospective study of lung cancer patients ation dose exceeding 5 Gy, 10 Gy, 20 Gy and 30 Gy, who underwent radiation therapy to assess the value of respectively) were calculated from the lung Dose Volume the cytokine dynamics, as well as of dosimetric factors, for Histogram (DVH). The lung dosimetric factors were calcu- predicting the risk of developing symptomatic RP. lated with subtraction of the GTV. The both lungs were considered either as a single paired organ or as two sepa- Methods rate organs. Patient eligibility We prospectively included lung cancer patients who Radiation therapy was delivered with 10 MV X-ray from a underwent radiation therapy from November 2006 to LINAC (Siemens, Concord, CA, U.S.A.). The median radi- April 2007 at Seoul Saint Mary's Hospital, Seoul, Korea. ation dose given was 55.6 Gy (range: 45-66 Gy) and the median fraction size was 1.8 Gy (range: 1.8-3 Gy). Patients eligible for this study were those who had histo- logically proven lung cancer, received three-dimensional Follow up and the definition of RP conformal radiation therapy (3D-CRT) above 45 Gy and The patients visited the hospital for follow-up at 2 weeks an ECOG performance status score of 0-2 with the life and 1, 2, 3 and 6 months after RT. Chest X-rays were per- expectancy of more than 6 months. Written informed formed at every follow-up visit and chest CT was done at consents were obtained from all of the patients in accord- 1, 3 and 6 months after RT. We evaluated the symptoms ance with the procedures of the Institutional Review and signs of RP and the radiological changes upto at least Board of the Hospital. 6 months after treatment. The end point of this study was the development of ≥ grade 2 RP. Grading of the RP was Clinical evaluation and treatment description recorded using the scale defined by RTOG, which is based History taking and physical examination that emphasized on the severity of clinical symptoms of patients with radi- the respiratory system were performed before, during and ographic changes. It is outlined as follows: Grade 1 pneu- after RT. All of the patients underwent pretreatment chest monitis is for asymptomatic patients or who have mild X-ray, chest CT scan, pulmonary function tests (PFT), symptoms(dry coughing or dyspnea on exertion) with bone scan and if necessary, PET CT and brain MRI. radiographic findings; Grade 2 pneumonitis is for patients Page 2 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 who are moderately symptomatic(persistent coughing For finally 34 evaluable patients, the histology was 14 requiring narcotic or antitussive); Grade 3 pneumonitis is squamous cell carcinomas, 6 adenocarcinomas, 11 small for severely symptomatic patients(severe coughing unre- cell carcinomas and 3 others. Twenty-five patients were sponsive to narcotic or antitussive or dyspnea at rest, male and nine patients were female. The median patient intermittent oxygen or steroid required); Grade 4 pneu- age was 63 (age range: 42-78). The ECOG performance monitis is for patients with severe respiratory insufficiency status was 0 or 1 for 29 patients (75%). Twenty-six who needs continuous oxygen or assisted ventilation; patients (78%) had a smoking history. Chemotherapy Grade 5 pneumonitis means death due to aggravation of was done before radiation in 21 patients (62%), and con- pneumonia. Lung injury in the immune suppressed host current chemoradiotherapy was done in 15 patients is associated with diversity of etiologies: sepsis, respiratory (44%). The used regimens for concurrent chemotherapy infection, irradiation, reperfusion injury, chemotherapeu- were combinations of etoposide and cisplatin (6 patients) tic agents and other drug reactions. Multidisciplinary or docetaxel and cisplatin (4 patients). Other chemother- Team of Lung Cancer in Seoul St. Mary's Hospital apeutic agents included weekly taxol (4 patients) or cispl- reviewed our patients suspected of having RP and atin (1 patient). Twelve patients (35.3%) had previous excluded other confirmed causes of pneumonia that mim- surgery for the lung cancer. The results of the pretreatment icked RP such as infectious pneumonia or disease progres- pulmonary function test for all the patients were as fol- sion. lows: the median FEV1 was 1.81 L (70%), the median FEV1/FVC 67% and the median DLCO 78%. The patients Analysis of the circulating cytokines characteristics are summarized in Table 1. Peripheral blood samples were collected from the patients at six points: before treatment and at the beginning, in the Radiation pneumonitis middle of and at the end of RT and at 2 and 4 weeks after The median duration of follow up was 10 months (range: the completion of RT. By using meticulous handling pro- 1 - 23 months). Of the 34 patients who were included in cedure, the blood samples were collected into an EDTA- the study, 17 patients (50%) developed any grade of RP. contained tube. Then within an hour, centrifugation at The severity of pneumonitis was grade 1 in 9 patients 3000 × g was carried out for 20 min at 4°C, the plasma (26.5%), grade 2 in 5 patients (14.7%) and grade 3 in 2 supernatant collected and stored in aliquots of 500 μL at - patients (5.9%). One patients (2.9%) died of aggravation 80°C until use. The plasma IL-1α, IL-6, IL-10, TNF-α and of RP. The median time to the onset of pneumonitis was TGF-β1 levels were determined by an ELISA kit (R&D sys- 1.6 months (range: 0-3.3 months). tems inc., Minneapolis, MN, U.S.A.). To minimize the var- iability of each ELISA assay, we completed all of the assay Among 8 patients who developed ≥2 RP, 7 patient with just 2 times and compared standard dose density received concurrent chemoradiation. The most frequently curve of different assays to ensure reproducibility. used regimens were combinations of etoposide and cispl- atin (4 patients) or docetaxel and cisplatin (2 patients). Statistical analysis Statistical analysis was performed to correlate radiation The patient who developed grade 5 pneumonitis was 74 pneumonitis with various potentially predictive parame- year-old ex-smoker. He was newly diagnosed for squa- ters (dosimetric variables and changes in cytokine levels). mous cell carcinoma. The stage was IIIB, so concurrent The differences in dosimetric variables and cytokine levels chemoradiation with weekly taxol was recommended between RP and non-RP groups were compared by con- considering his good performance status (ECOG1) and ducting the student's t-test. Analysis of variance for PFTs (FEV1 = 2.52L, DLCO = 84%). He was relatively well repeated measures was used to examine the interaction tolerable until the end of treatment. However, he had between changes of cytokine levels during time course coughing, progressive dyspnea and fever with new devel- and RP occurrence. All the calculations were performed oped consolidation corresponding to radiation field after using the SAS system (SAS Institute Inc., Cary, NC, USA). 1.5 months completion of RT. No specific infectious cause A two-sided value of < 0.05 was considered statistically was identified. He received intravenous steroid therapy significant. and ventilator care, but died despite of 4 weeks of inten- sive care. Results Patient Characteristics Dosimetric parameters Forty-three patients were initially included in the study, All the dosimetric factors were analyzed for the lung both but nine patients were then excluded from analysis due to as a paired organ and as a separate organ. Most of the dosi- incomplete treatment (2 patients), unsatisfactory blood metric factors showed an association with RP. The data are sampling (3 patients) and follow-up loss or a follow-up shown in Table 2. For the lung as a paired organ, V5, V10 that was less than 6 months (4 patients). and V20 were statistically significant factors for the occur- Page 3 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 1: Patients characteristics (n = 34) Characteristics Number of patients (%) Gender Male 25 (74) Female 9 (26) Age(yrs) Median 63 Range 42~78 ECOG 0 13 (38) 1 16 (47) 2 5 (15) Site Upper 22 (65) Lower 10 (29) Upper+Lower 2 (6) Histology Sqamous cell ca 14 (41) Adeno ca 6 (18) Small cell ca 11 (32) Other 3 (9) Location Central 26 (76) Peripheral 8 (24) Previous surgery No 22 (65) Yes 12 (35) Previous chemotherapy No 13 (38) Yes 21 (62) Concurrent chemotherapy No 19 (56) Yes 15 (44) Smoking Never 7 (21) Previous 13 (39 Current 13 (39) PFTs FEV1(L) Median 1.8 Range 0.8~4.5 FEV1/FVC(%) Median 67 Range 34~94 DLCO(%) Median 78 Range 53~133 Abbreviations: ECOG, Eastern Cooperative Oncology Group Performance Scale; PFTs, Pulmonary Function Tests rence of RP (p = 0.035, p = 0.049, and p = 0.049, respec- 0.062). However, TGF-β1 level of patients who did not tively). The V30 and MLD were marginal significant (p = develop RP began to decrease relative to their pretreat- 0.068, p = 0.077). For the lung as a separate organ, the ment level after middle of RT. MLD, V5, V10 and V20 values were statistically significant factors for the occurrence of RP (p = 0.018, p = 0.003, p = The pretreatment TGF-β1 level of the patient who died of pg/ml). The TGF-β1 level 0.006 and p = 0.032, respectively). The V30 was margin- RP was relatively high (2.9 × 10 ally significant (p = 0.053). decreased during RT as same as other RP patients. His TGF-β1 level at 4 weeks after RT was markedly increased Serum cytokine levels upto 4.2 × 10 pg/ml. TGF- 1 As demonstrated in Table 3, the mean pretreatment TGF- We performed an ANOVA model for repeated-measures β1 level was 2.8 ± 0.8 × 10 pg/ml for the RP group and for analysis of chronological change in TGF-β1 level and 2.3 ± 1.1 × 10 pg/ml for the non-RP group. The patients found that there were significant associations between the who developed pneumonitis showed a higher level of pre- changes of TGF-β1 level during the time course of radia- treatment TGF-β1, but this was not statistically significant tion and the risk of developing RP (p < 0.001 for develop- (p = 0.157). During the period of radiation treatment, ment of any grade of RP, p < 0.0001 for development of from the beginning of RT to the middle of RT, the RP grade ≥2 RP). These chronological changes in the serial group tended to show a decrease in the TGF-β1 level. TGF-β1 levels are demonstrated in Table 4, Fig 1. However, the TGF-β1 level began to increase at the end of RT in the RP group and became significantly higher at 4 IL-6 weeks after RT (p = 0.007). These elevation of TGF-β1 level The pretreatment IL-6 level was higher in the non-RP after RT were same in the patients of ≥ grade 2 RP (p = group compared to the RP group (49.1 ± 123.3 pg/ml vs Page 4 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 2: Dosimetric risk factors for development of RP ≥ grade 2 Mean ± SD DVH parameters RP Non-RP p value* Whole lung MLD (cGy) 1188 ± 354 957 ± 382 0.077 V5 (%) 58 ± 15 45 ± 12 0.035 V10 (%) 43 ± 13 34 ± 13 0.049 V20 (%) 29 ± 8 22 ± 11 0.049 V30 (%) 22 ± 7 17 ± 9 0.068 V40 (%) 15 ± 5 12 ± 7 0.299 Ipsilateral lung MLD (cGy) 1986 ± 491 1512 ± 612 0.018 V5 (%) 78 ± 13 63 ± 17 0.003 V10 (%) 67 ± 11 52 ± 19 0.006 V20 (%) 49 ± 11 38 ± 17 0.032 V30 (%) 38 ± 11 29 ± 15 0.053 V40 (%) 24 ± 9 22 ± 14 0.655 Abbreviations: RP, radiation pneumonitis; DVH, dose volume histogram; MLD, mean lung dose; V5, the percentage of the irradiated lung volume receiving a radiation dose exceeding 5 Gy; V10, the percentage of the irradiated lung volume receiving a radiation dose exceeding 10 Gy; V20, the percentage of the irradiated lung volume receiving a radiation dose exceeding 20 Gy; V30, the percentage of the irradiated lung volume receiving a radiation dose exceeding 30 Gy; V40, the percentage of the irradiated lung volume receiving a radiation dose exceeding 40 Gy * student's t-test 16.7 ± 23.9 pg/ml, respectively, p = 0.387). However, the Other cytokines IL-6 level between each patient showed wide variation The IL-1α, IL-10 and TNF-α level were stable during the compared to the TGF-β1 level, which showed a relatively whole course of RT and they were not correlated with the stable range of variation. The IL-6 level was low before risk of RP (data not shown). and during RT, but it began to increase after RT in both groups. The changes of the IL-6 during the time course Discussion were similar whether patients developed RP or not. Our Being aware of the risk of radiation pulmonary injury is an data showed a wide range of variation in the circulatory important aspect of patient management in the era of IL-6 levels, but relatively little changes with the develop- combination of chemoradiation for treating lung cancer. ment of RP. The recent studies on the new effective chemotherapeutic agents or hyperfractionated RT have reported that the inci- Table 3: Changes of mean TGF-β1 level during the course of dence of grade ≥ 3 RP in patients who underwent concur- radiation therapy rent chemoradiotherapy is between 21-23%[6,7]. Along Mean ± SD (× 10 pg/ml) with combined chemotherapy, other patient-specific or Time RP Non-RP p value* treatment-specific factors have been identified as predic- tors of developing RP. However, currently, there are no RP1 Before RT 2.8 ± 0.8 2.3 ± 1.1 0.157 generally accepted methods available to accurately predict (n = 17) Beginning of RT 2.4 ± 0.7 2.5 ± 1.1 0.738 an individual patient's risk of developing RT-induced pul- Middle of RT 2.0 ± 0.6 2.4 ± 1.2 0.19 monary morbidity. The purpose of this study was to assess End of RT 2.1 ± 0.7 2.2 ± 0.9 0.879 the values of the cytokine dynamics and dosimetric factors 2 wks after RT 2.1 ± 0.4 2.0 ± 0.8 0.806 to predict the risk of developing symptomatic RP. 4 wks after RT 2.8 ± 1.1 1.7 ± 0.5 0.007 RP2 Before RT 2.5 ± 1.0 2.6 ± 0.9 0.844 The dose of irradiation administered to patients is distrib- (n = 8) Beginning of RT 1.9 ± 0.6 2.6 ± 0.9 0.079 uted in a 3-D volume on the DVH. According to a litera- Middle of RT 1.8 ± 0.7 2.3 ± 0.9 0.168 ture review, dosimetric factors such as V5, V10, V20, V30, End of RT 1.9 ± 0.6 2.2 ± 0.8 0.352 V (effective volume) and MLD have statistically signifi- eff 2 wks after RT 2.1 ± 0.7 2.0 ± 0.9 0.787 cant correlation to symptomatic radiation pneumonitis 4 wks after RT 3.3 ± 1.7 2.0 ± 0.9 0.062 [9-11]. Also in the current study, most of the dosimetric factors showed an association with RP. Because the lung Abbreviations: RT, radiation therapy; wks, weeks; RP1, development of any grade of radiation pneumonitis; RP2, development of grade ≥ 2 function may not be uniform across all regions of the radiation pneumonitis; lung, it is unlikely that such a simplistic dose-volume rela- * student's t-test Page 5 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 4: Results of repeated measures ANOVA about the changes of mean TGF-β1 level during the course of radiation therapy Mean ± SD (× 10 pg/ml) Repeated measures ANOVA Before RT Beginning of Middle of RT End of RT 2 wks after RT 4 wks after RT Source F p value RT RP1 2.8 ± 0.8 2.4 ± 0.7 2.0 ± 0.6 2.1 ± 0.7 2.1 ± 0.4 2.8 ± 1.1 RP1 0.28 0.598 (n = 17) Non-RP 2.3 ± 1.1 2.5 ± 1.1 2.4 ± 1.2 2.2 ± 0.9 2.0 ± 0.8 1.7 ± 0.5 time 2.36 0.043 (n = 17) time*RP1 4.79 0.001 RP2 2.5 ± 1.0 1.9 ± 0.6 1.8 ± 0.7 1.9 ± 0.6 2.1 ± 0.7 3.3 ± 1.7 RP2 0.05 0.828 (n = 8) Non-RP 2.6 ± 0.9 2.6 ± 0.9 2.3 ± 0.9 2.2 ± 0.8 2.0 ± 0.9 2.0 ± 0.9 time 3.24 0.008 (n = 26) time*RP2 6.27 0.0001 Abbreviations: RT, radiation therapy; wks, weeks; RP1, development of any grade of radiation pneumonitis; RP2, development of grade ≥ 2 radiation pneumonitis tionship exists. However, these parameters are easy to cal- oped symptomatic pneumonitis [14]. Kong et al. reported culate and useful in the clinical setting. that TGF-β1 decreased during RT in patients with an increased pretreatment plasma TGF-β1 level, yet this Although dosimetric factors are important, these factors didn't normalize even by the completion of treat- don't take into consideration the molecular biological ment[19]. Other clinical studies have reported that the events that may be responsible for radiation-induced absolute level or the relative ratio of the TGF-β1 level responses of normal tissue. Molecular events have been showed meaningful changes during or after RT in patients shown to occur much earlier than the clinically apparent who suffered RP. However, the correlated time points radiation responses. Exposure to ionizing radiation rap- were different in each study and the patterns of the idly triggers a cascade of genetic and molecular events, changes were not always same [14,18-20]. We also evalu- which is an active process involving the production of a ated the TGF-β1 ratios (the ratios of a value from a partic- number of inflammatory and fibrogenic cytokines by var- ular time-point divided by the pre-RT value) presented in ious cells in lung, for example, macrophages, epithelial table 5. In the patients who experienced grade ≥2 RP, the cells, endothelial cells, pneumocytes and fibroblasts. TGF-β1 ratios tended to show a decrease than that of the These molecular processes are perpetuated beyond the non-RP group, from the beginning of RT to the end of RT. time point at which the acute insult has been removed. However, the TGF-β1 level began to increase at the 2 Several recent studies have shown that cytokines (IL-1, IL- weeks after RT in the RP group and became more higher 6, IL-8, IL-10, TNF-α, platelet-derived growth factor and at 4 weeks after RT (1.6 ± 1.0 vs 0.8 ± 0.3, p = 0.081). The TGF-β), surfactant apoproteins and cell adhesion mole- time point predictive of RP is different from the data of cules (ICAM-1, E-selectin) have important roles in RT- Zhao et al which was 4 weeks during 6 weeks' course of RT induced pulmonary injury [8,9,12-18]. A study of the [20]. A recent report from Zhao et al suggests the combi- dynamics of the serum cytokines during the early course nation of TGF-β1 and MLD may help stratify the patients of the treatment would be useful for predicting the risk of for their risk of RP to improve the predictive power [21]. pulmonary injury and for the early intervention of pneu- Their data show the incidence of RP was 4.3% in patients monitis. Because these cytokines are thought to be key with a TGF-β1 ratio ≤ 1 and MLD ≤ 20 Gy, and 66.7% in mediators of lung toxicity, many of them have been exam- those with a TGF-β1 ratio ≥ 1 and MLD ≥ 20 Gy. ined as potential early markers for radiation pneumonitis. The TGF-β1 levels in bronchoalveolar lavage(BAL) fluid Profibrogenic cytokine TGF-β1 is the most extensively from the irradiated area increased continuously during investigated among the various biological markers in radi- and after RT compared to the pretreatment levels in the RP ation-induced injuries. We observed that TGF-β1 group[17]. On the other hand, in several studies the pat- decreased during RT and it began to increase at the end of tern of the changes of the TGF-β1 level was not distinct RT in the patients who developed RP. There have been between the RP and non-RP groups [12,15]. The reasons similar reported results for the TGF-β1 dynamics in rela- for such conflicting results may be explained by the fact tion to the development of pneumonitis after radiother- that numerous factors can falsely increase TGF-β1 levels apy for lung cancer. Hur et al. found that TGF-β1 was and confound their predictive value for RP occurrence. decreased during RT and it was markedly increased at 2-4 First, the tumor stroma may be responsible for the pro- weeks after the completion of RT for patients who devel- duction of TGF-β1 in lung cancer patients. The mean TGF- Page 6 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 [15,17]. In this study, the IL-6 level was low before and during RT, but it began to increase after RT. These findings are similar those of Chen et al [24]. However, the changes of the IL-6 level were similar whether patients developed RP or not. This suggests that the IL-6 produced by the lung is not a major determinant of the circulating IL-6 levels. The mean IL-6 concentrations were significantly higher in the lung cancer patients than in the normal controls, and the patients with metastatic tumor had higher IL-6 levels than those patients with undisseminated disease, suggest- ing that neoplastic cells may produce IL-6 [25]. IL-6 is an acute phase inflammatory cytokine, and this would sug- gest that the measurement of circulating IL-6 can reflect the inflammatory state of the lung. The IL-6 levels fre- quently increase in patients suffering with several pulmo- nary diseases, including infectious pneumonia, interstitial pneumonia and chronic obstructive pulmonary dis- Th RP g Figure 1 e p rou attern of the ch p and the non-RP gr anges of the mean TGF- oup β1 level in the ease[17]. The pattern of the changes of the mean TGF-β1 level in the RP group and the non-RP group. The TGF-β1 There are other potential biological predictors of RP, but level began to increase at the end of RT in the RP group and none has been conclusively demonstrated to identify the it became significantly higher at 4 weeks after RT (p = 0.007). patients who are at a high risk of radiation-induced pul- However TGF-β1 level of patients who did not develop RP monary toxicity [9,12,15,16]. Host-associated diseases, began to decrease relative to their pretreatment level after the proportion of gross disease at the time of irradiation middle of RT. The solid line shows the mean level of TGF-β1 and pre-RT treatment such as chemotherapy are all associ- in the RP group and the dashed line shows the mean level of TGF-β1 in the non-RP group. The data are presented as ated with local cytokine production [5,15]. The addition mean ± standard error of the mean. of radiation on a background of subclinical damage may augment a cytokine cascade and increase the severity of acute and late side effects. β1 level in lung cancer patients was higher than that in the normal controls (p < 0.001). According to a pathology Although various cytokines are important in the patho- slide review, the degree of fibrosis that is present in tumor genesis of radiation-induced pulmonary injury, our is also significantly correlated with an elevated plasma results show that the changes of TGF-β1 could be an inde- TGF-β1 level (p = 0.03). The dynamics of the plasma TGF- pendent predictor of developing RP and well correlated β1 have been suggested to be a marker of RT-induced nor- with the time course of radiation therapy. However, the mal tissue injury as well as a marker of tumor response. role of cytokine markers in the cytokine cascades that pro- After radiotherapy, the patients who were alive with dis- mote pulmonary injury deserves further investigation. ease had significantly higher TGF-β1 levels than those Also, treatment strategies designed to block this patho- who were alive with no evidence of disease (p = 0.02) [19] logic process may need to be continued well beyond the Second, careful handling of the sample is also important. completion of RT. Variations of the centrifugation conditions and platelet contamination could artificially elevate plasma TGF-β1 Table 5: Changes of mean TGF-β1 ratio during the course of level[19,22] radiation therapy Mean of Ratio ± SD IL-6 is a pleiotropic inflammatory cytokine that is impor- Time RP Non-RP p value* tant in regulating immunologic and inflammatory responses. IL-6 levels before, during and after thoracic RT RP2 Before RT were significantly higher in those patients who developed (n = 8) Beginning of RT 0.8 ± 0.2 1.0 ± 0.2 0.019 pneumonitis in several reports [13,23]. Especially Arpin et Middle of RT 0.8 ± 0.4 0.9 ± 0.4 0.238 al reported covariation of proinflammatory cytokine (IL- End of RT 0.8 ± 0.2 0.9 ± 0.3 0.346 2 wks after RT 1.0 ± 0.6 0.9 ± 0.3 0.576 6) and anti-inflammatory cytokine (IL-10) levels during 4 wks after RT 1.6 ± 1.0 0.8 ± 0.3 0.081 the first 2 week of RT were independent predictive evi- dence of RP. However, other studies have failed to find a Abbreviations: RT, radiation therapy; wks, weeks; RP2, development relationship between IL-6 and the radiation-induced pul- of grade ≥ 2 radiation pneumonitis monary symptoms, like what our results have shown * student's t-test Page 7 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 small-cell lung cancer: radiation therapy oncology group Conclusion protocol 91-06. J Clin Oncol 1996, 14:1055-1064. Although the current study had a limited number of 7. Kouroussis C, Mavroudis D, Kakolyris S, Voloudaki A, Kalbakis K, patients, we demonstrated that, in the patients who devel- Souglakos J, Agelaki S, Malas K, Bozionelou V, Georgoulias V: High incidence of pulmonary toxicity of weekly docetaxel and oped RP, the TGF-β1 level decreased during RT and began gemcitabine in patients with non-small cell lung cancer: to increase at the end of RT. And the TGF-β1 level at 4 results of a dose-finding study. Lung cancer 2004, 44:363-368. 8. Claude L, Perol D, Ginestet C, Falcheroc L, Arpind D, Vincente M, weeks after RT was significantly higher than the TGF-β1 Martela I, Hominalf S, Cordierg JF, Carrie C: A prospective study level of the patients who didn't develop RP. TGF-β1 may on radiation pneumonitis following conformal radiation contribute to the process leading to a radiation injury in therapy in non-small-cell lung cancer: clinical and dosimetric factors analysis. Radiother Oncol 2004, 71:175-181. human lung tissue. While the change of TGF-β1 level did 9. Wang S, Liao Z, Wei X, Liu HH, Tucker SL, Hu CS, Mohan R, Cox JD, not take place early in time course of RT with having a pre- Komaki R: Analysis of clinical and dosimetric factors associ- dictive value in our study, the incorporation of the biolog- ated with treatment-related pneumonitis (TRP) in patients with non-small-cell lung cancer(NSCLC) treated with con- ical parameters into the dosimetric data that have been current chemotherapy and three-dimensional conformal developed to predict radiation-induced lung injury may radiotherapy(3D-CRT). Int J Radiat Oncol Biol Phys 2006, 66:1399-1407. improve the predictive accuracy. Further research must 10. Vujaskovic Z, Marks LB, Anscher MS: The physical parameters continue to identify biomarkers that will one day allow us and molecular events associated with radiation-induced lung to tailor our therapies in response to the highly accurate toxicity. Semin Radiat Oncol 2000, 10:296-307. 11. Hartsell WF, Scott CB, Dundas GS, Mohiuddin M, Meredith RF, Rubin predictions of risk for the development of radiation pneu- P, Weigensberg IJ: Can serum markers be used to predict acute monitis. and late toxicity in patients with lung cancer ? analysis of RTOG 91-03. Am J Clin Oncol 2007, 30:368-376. 12. Chen Y, Williams J, Ding I, Hernady E, Liu W, Smudzin T, Finkelstein Competing interests JN, Rubin P, Okunieff P: Radiation pneumonitis and early circu- The authors declare that they have no competing interests. latory cytokine markers. Semin Radiat Oncol 2002, 12(Suppl 1):26-33. 13. Chen Y, Rubin P, Williams J, Hernady E, Smudzin T, Okunieff P: Cir- Authors' contributions culating IL-6 as a predictor of radiation pneumonitis. Int J JYK performed the collection of blood samples, acquisi- Radiat Oncol Biol Phys 2001, 49:641-648. 14. Hur WJ, Yoon SM, Lee HS, Yang KM, Shin GH, Son CH, Han JY, Lee tion of clinical data and drafted the manuscript. YSK KN, Jeong MH: The measurements of plasma cytokines in designed and coordinated the study, checked statistical radiation-induced pneumonitis in lung cancer patients. J Korean Soc Ther Oncol 2000, 18:314-320. results, read and edited the manuscript. YKK coordinated 15. Hart JP, Broadwater G, Rabbani Z, Moeller BJ, Clough R, Huang D, and performed laboratory work. HJP interpreted radiolog- Sempowsk GA, Dewhirst M, Pizzo SV, Vujaskovic Z, Anscher MS: ical findings. SJK, JHK, YPW, SCY, SNL and HSJ performed Cytokine profiling for prediction of symptomatic radiation- induced lung injury. Int J Radiat Oncol Biol Phys 2005, 63:1448-1454. evaluation of patients and read the manuscript. All the 16. Ishii Y, Kimura S: Soluble intercellular adhesion molecule-1 as authors read and approved the final manuscript. an early detection marker for radiation pneumonitis. Eur Respir J 1999, 13:733-138. 17. Barthelemy-Brichant N, Bosquee L, Cataldo D, Corhay JL, Gustin M, Acknowledgements Seidel L, Thiry A, Ghaye B, Nizet M, Albert A, Deneufbourg JM, Bar- The authors wish to acknowledge the financial support of the Catholic tsch P, Nusgens B: Increased IL-6 and TGF-β1 concentrations Medical Center Research Foundation made in the program year of 2006. in bronchoalveolar lavage fluid associated with thoracic radi- otherapy. Int J Radiat Oncol Biol Phys 2004, 58:758-767. 18. Novakova-Jiresova A, Gameren MM, Coppes RP, Kampinga HH, References Groen H: Transforming growth factor-β plasma dynamics and 1. Willner J, Caragiani E, Tschammler A, Flentje M: Dose, volume, and post-irradiation lung injury in lung cancer patients. Radiother tumor control prediction in primary radiotherapy of non- Oncol 2004, 71:183-189. small-cell lung cancer. Int J Radiat Oncol Biol Phys 2002, 19. Kong FM, Washington MK, Jirtle RL, Anscher MS: Plasma trans- 52:382-389. forming growth factor-β1 reflects disease status in patients 2. Kong FM, Ten Haken RK, Schipper MJ, Sullivan MA, Chen M, Lopez with lung cancer after radiotherapy: a possible tumor C, Kalemkerian GP, Hayman JA: High-dose radiation improved marker. Lung Cancer 1996, 16:47-59. local tumor control and overall survival in patients with inop- 20. Zhao L, Sheldon K, Chen M, Yin MS, Hayman JA, Kalemkerian GP, erable/unresectable non-small-cell lung cancer: Long-term Arenberg D, Lyons SE, Curtis JL, Davis M, Cease KB, Brenner D, results of a radiation dose escalation study. Int J Radiat Oncol Anscher MS, Lawrence TS, Kong FM: The predictive role of Biol Phys 2005, 63:324-333. plasma TGF-beta1 during radiation therapy for radiation- 3. Marino P, Preatoni A, Cantoni A: Randomized trials of radiother- induced lung toxicity deserves further study in patients with apy alone versus combined chemotherapy and radiotherapy non-small cell lung cancer. Lung Cancer 2008, 59:232-239. in stages IIIa and IIIb non-small cell lung cancer a meta-anal- 21. Zhao L, Wang L, Ji W, Wang X, Zhu X, Hayman JA, Kalemkerian GP, ysis. Cancer 1995, 76:593-601. Yang W, Brenner D, Lawrence TS, Kong FM: Elevation of plasma 4. Chevalier TL, Arriagada R, Quoix E, Ruffie P, Martin M, Douillar JY, TGF-beta1 during radiation therapy predicts radiation- Tarayre M, Lacombe-Terrier MJ, Laplanche A: Radiotherapy alone induced lung toxicity in patients with non-small-cell lung can- versus combined chemotherapy and radiotherapy in unre- cer: a combined analysis from Beijing and Michigan. Int J sectable non-small cell lung carcinoma. Lung Cancer 1994, Radiat Oncol Biol Phys 2009, 74:1385-1390. 10(Suppl 1):S239-S244. 22. Wakefield LM, Letterio J, Chen J, Danielpour D, Allison RS, Pai LH, 5. Mehta V: Radiation pneumonitis and pulmonary fibrosis in Denicoff AM, Noone MH, Cowan KH, O'Shaughnessy JA: Trans- non-small-cell lung cancer: Pulmonary function, prediction, forming growth factor-beta1 circulates in normal human and prevention. Int J Radiat Oncol Biol Phys 2005, 63:5-24. plasma and is unchanged in advanced metastatic breast can- 6. Lee JS, Komaki R, Fossella FV, Dundas GS, McDonald S, Byhardt RW, cer. Clin Cancer Res 1995, 1:129-136. Curran WJ Jr: Concurrent chemoradiation therapy with oral 23. Arpin D, Perol D, Blay JY, Falchero L, Claude L, Vuillermoz-Blas S, etoposide and cisplatin for locally advanced inoperable non- Martel-Lafay I, Ginestet C, Alberti L, Nosov D, Etienne-Mastroianni B, Page 8 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Cottin V, Perol M, Guerin JC, Cordier JF, Carrie C: Early variations of circulating interleukin-6 and interleukin-10 levels during thoracic radiotherapy are predictive for radiation pneumo- nitis. J Clin Oncol 2005, 23:8748-8756. 24. Chen Y, Hyrien O, William J, Okunieff P, Smudzin T, Rubin P: Inter- lekin (IL)-1A and IL-6: Application to the predictive diagnos- tic testing of radiation pneumonitis. Int J Radiat Oncol Biol Phys 2005, 62:260-266. 25. De Vita F, Orditura M, Auriemma A, Infusino S, Roscigno A, Catalano G: Serum levels of interleukin-6 as a prognostic factor in advanced non-small cell cancer. Oncology reports 1998, 5:649-652. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 9 of 9 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Oncology Springer Journals

The TGF-β1 dynamics during radiation therapy and its correlation to symptomatic radiation pneumonitis in lung cancer patients

Loading next page...
 
/lp/springer-journals/the-tgf-1-dynamics-during-radiation-therapy-and-its-correlation-to-YiBURHiGan
Publisher
Springer Journals
Copyright
Copyright © 2009 by Kim et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Oncology; Radiotherapy
eISSN
1748-717X
DOI
10.1186/1748-717X-4-59
pmid
19943923
Publisher site
See Article on Publisher Site

Abstract

Backgroud: The underlying molecular and cellular mechanisms of radiation pneumonitis (RP) are very complex. Several biological factors need to be considered together with the well known dosimetric parameters for understanding the molecular events in developing RP in lung cancer patients. The aim of this study was to correlate the variations of the cytokine levels in lung cancer patients during radiation therapy (RT) with the occurrence of symptomatic RP. Methods: Thirty-four lung cancer patients who received three-dimensional conformal radiation therapy were evaluated prospectively. Serial blood samples before, at the beginning, in the middle of, at the end of RT and 2 and 4 weeks after RT were analyzed for IL-1α, IL-6, IL-10, TNF-α and TGF-β1 by performing enzyme-linked immunosorbent assay. The predictive values of dosimetric factors for RP were evaluated, too. Results: Overall, 8 patients (23.5%) had grade ≥ 2 RP. By serial measurement of cytokines level, only the TGF-β1 level showed a correlation to the symptomatic RP. None of the other cytokines, IL-1α, IL-6, IL-10 and TNF-α level was correlated with the risk of RP. The mean pretreatment TGF- β1 level did not differ between RP and non-RP groups. However, during the period of radiation treatment, the TGF-β1 level began to increase at the end of RT in the RP group and became significantly higher 4 weeks after RT (p = 0.007). Using an ANOVA model for repeated-measures, we found significant associations between the changes of TGF-β1 during the time course of the RT and the risk of developing RP (p < 0.001). Most of the dosimetric factors showed a significant association with RP. Conclusion: Our results show that the changes of TGF-β1 could be correlated with RP and the incorporation of the biological parameters into the dosimetric data could be useful for predicting symptomatic RP. Page 1 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 A chest CT scan including the entire lung volume was per- Background Increasing radiation dose and combining it with chemo- formed with the immobilization device at 3 mm scan therapy were found to improve the local control for lung thickness. The structures of interest, such as gross tumor cancer and thus improve the overall survival of lung can- volume (GTV), clinical target volume (CTV), and normal cer patients [1-4]. However, the tolerance of the surround- structures (esophagus, spinal cord, heart, and lung) were ing normal tissues to radiation therapy limits the level of defined and contoured on the multiple CT images. The dose that can be delivered for the treatment of lung can- gross tumor volume included the tumor and clinically cer. involved nodes. GTV was determined by chest CT and/or PET-CT. CTV included GTV with additional uniform 5 cm The risk of radiation pulmonary toxicity may increase for expansion and ipsilateral hilar lymph nodes. The unin- the following patients: patients with a poor performance volved mediastinal, and supraclavicular nodal regions status or inadequate pulmonary function, patients who were not routinely included in the CTV, but high risk undergo combined chemoradiotherapy and patients nodal stations near GTV nodes were included occasion- receiving an increased total radiation dose and treatment ally. The PTV was obtained by a 0.7~1.0 cm margin 3-D volume. Dosimetric parameters such as V5, V10, V20, expansion from the CTV. An additional 0.5~1.0 cm mar- V30, V40, the mean lung dose (MLD) and the normal tis- gin was added to superior and inferior direction to sue complications probability (NTCP) have been reported account for respiratory motion as assessed under fluoros- to be related with RP by several retrospective or prospec- copy. The beam arrangement was planned to minimize tive studies [5-11]. However, the underlying molecular the irradiated lung volume usually using 3 to 5 coplanar and cellular mechanisms of RP are very complex. Several oblique beams. Some patients of having bulky tumor biological factors need to be considered together with the were treated with AP/PA fields to include the CTV for the above mentioned dosimetric parameters for understand- first 30~40 Gy, followed by off-cord oblique beam to the ing the molecular events in developing radiation-induced GTV usually composed of 3~5 beams. After making cor- complications in normal tissue [5,8,12]. Several clinical rection for the tissue inhomogeneity using convolution/ reports have suggested a possible role of profibrogenic superposition algorithm of the treatment planning system and proinflammatory cytokines in the modulation of (Pinnacle, Philips Medical System, Andover, MA, U.S.A.), radiation pulmonary injury [12-18]. the mean lung dose (MLD), V5, V10, V20 and V30 (the percentages of the irradiated lung volume receiving a radi- We designed a prospective study of lung cancer patients ation dose exceeding 5 Gy, 10 Gy, 20 Gy and 30 Gy, who underwent radiation therapy to assess the value of respectively) were calculated from the lung Dose Volume the cytokine dynamics, as well as of dosimetric factors, for Histogram (DVH). The lung dosimetric factors were calcu- predicting the risk of developing symptomatic RP. lated with subtraction of the GTV. The both lungs were considered either as a single paired organ or as two sepa- Methods rate organs. Patient eligibility We prospectively included lung cancer patients who Radiation therapy was delivered with 10 MV X-ray from a underwent radiation therapy from November 2006 to LINAC (Siemens, Concord, CA, U.S.A.). The median radi- April 2007 at Seoul Saint Mary's Hospital, Seoul, Korea. ation dose given was 55.6 Gy (range: 45-66 Gy) and the median fraction size was 1.8 Gy (range: 1.8-3 Gy). Patients eligible for this study were those who had histo- logically proven lung cancer, received three-dimensional Follow up and the definition of RP conformal radiation therapy (3D-CRT) above 45 Gy and The patients visited the hospital for follow-up at 2 weeks an ECOG performance status score of 0-2 with the life and 1, 2, 3 and 6 months after RT. Chest X-rays were per- expectancy of more than 6 months. Written informed formed at every follow-up visit and chest CT was done at consents were obtained from all of the patients in accord- 1, 3 and 6 months after RT. We evaluated the symptoms ance with the procedures of the Institutional Review and signs of RP and the radiological changes upto at least Board of the Hospital. 6 months after treatment. The end point of this study was the development of ≥ grade 2 RP. Grading of the RP was Clinical evaluation and treatment description recorded using the scale defined by RTOG, which is based History taking and physical examination that emphasized on the severity of clinical symptoms of patients with radi- the respiratory system were performed before, during and ographic changes. It is outlined as follows: Grade 1 pneu- after RT. All of the patients underwent pretreatment chest monitis is for asymptomatic patients or who have mild X-ray, chest CT scan, pulmonary function tests (PFT), symptoms(dry coughing or dyspnea on exertion) with bone scan and if necessary, PET CT and brain MRI. radiographic findings; Grade 2 pneumonitis is for patients Page 2 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 who are moderately symptomatic(persistent coughing For finally 34 evaluable patients, the histology was 14 requiring narcotic or antitussive); Grade 3 pneumonitis is squamous cell carcinomas, 6 adenocarcinomas, 11 small for severely symptomatic patients(severe coughing unre- cell carcinomas and 3 others. Twenty-five patients were sponsive to narcotic or antitussive or dyspnea at rest, male and nine patients were female. The median patient intermittent oxygen or steroid required); Grade 4 pneu- age was 63 (age range: 42-78). The ECOG performance monitis is for patients with severe respiratory insufficiency status was 0 or 1 for 29 patients (75%). Twenty-six who needs continuous oxygen or assisted ventilation; patients (78%) had a smoking history. Chemotherapy Grade 5 pneumonitis means death due to aggravation of was done before radiation in 21 patients (62%), and con- pneumonia. Lung injury in the immune suppressed host current chemoradiotherapy was done in 15 patients is associated with diversity of etiologies: sepsis, respiratory (44%). The used regimens for concurrent chemotherapy infection, irradiation, reperfusion injury, chemotherapeu- were combinations of etoposide and cisplatin (6 patients) tic agents and other drug reactions. Multidisciplinary or docetaxel and cisplatin (4 patients). Other chemother- Team of Lung Cancer in Seoul St. Mary's Hospital apeutic agents included weekly taxol (4 patients) or cispl- reviewed our patients suspected of having RP and atin (1 patient). Twelve patients (35.3%) had previous excluded other confirmed causes of pneumonia that mim- surgery for the lung cancer. The results of the pretreatment icked RP such as infectious pneumonia or disease progres- pulmonary function test for all the patients were as fol- sion. lows: the median FEV1 was 1.81 L (70%), the median FEV1/FVC 67% and the median DLCO 78%. The patients Analysis of the circulating cytokines characteristics are summarized in Table 1. Peripheral blood samples were collected from the patients at six points: before treatment and at the beginning, in the Radiation pneumonitis middle of and at the end of RT and at 2 and 4 weeks after The median duration of follow up was 10 months (range: the completion of RT. By using meticulous handling pro- 1 - 23 months). Of the 34 patients who were included in cedure, the blood samples were collected into an EDTA- the study, 17 patients (50%) developed any grade of RP. contained tube. Then within an hour, centrifugation at The severity of pneumonitis was grade 1 in 9 patients 3000 × g was carried out for 20 min at 4°C, the plasma (26.5%), grade 2 in 5 patients (14.7%) and grade 3 in 2 supernatant collected and stored in aliquots of 500 μL at - patients (5.9%). One patients (2.9%) died of aggravation 80°C until use. The plasma IL-1α, IL-6, IL-10, TNF-α and of RP. The median time to the onset of pneumonitis was TGF-β1 levels were determined by an ELISA kit (R&D sys- 1.6 months (range: 0-3.3 months). tems inc., Minneapolis, MN, U.S.A.). To minimize the var- iability of each ELISA assay, we completed all of the assay Among 8 patients who developed ≥2 RP, 7 patient with just 2 times and compared standard dose density received concurrent chemoradiation. The most frequently curve of different assays to ensure reproducibility. used regimens were combinations of etoposide and cispl- atin (4 patients) or docetaxel and cisplatin (2 patients). Statistical analysis Statistical analysis was performed to correlate radiation The patient who developed grade 5 pneumonitis was 74 pneumonitis with various potentially predictive parame- year-old ex-smoker. He was newly diagnosed for squa- ters (dosimetric variables and changes in cytokine levels). mous cell carcinoma. The stage was IIIB, so concurrent The differences in dosimetric variables and cytokine levels chemoradiation with weekly taxol was recommended between RP and non-RP groups were compared by con- considering his good performance status (ECOG1) and ducting the student's t-test. Analysis of variance for PFTs (FEV1 = 2.52L, DLCO = 84%). He was relatively well repeated measures was used to examine the interaction tolerable until the end of treatment. However, he had between changes of cytokine levels during time course coughing, progressive dyspnea and fever with new devel- and RP occurrence. All the calculations were performed oped consolidation corresponding to radiation field after using the SAS system (SAS Institute Inc., Cary, NC, USA). 1.5 months completion of RT. No specific infectious cause A two-sided value of < 0.05 was considered statistically was identified. He received intravenous steroid therapy significant. and ventilator care, but died despite of 4 weeks of inten- sive care. Results Patient Characteristics Dosimetric parameters Forty-three patients were initially included in the study, All the dosimetric factors were analyzed for the lung both but nine patients were then excluded from analysis due to as a paired organ and as a separate organ. Most of the dosi- incomplete treatment (2 patients), unsatisfactory blood metric factors showed an association with RP. The data are sampling (3 patients) and follow-up loss or a follow-up shown in Table 2. For the lung as a paired organ, V5, V10 that was less than 6 months (4 patients). and V20 were statistically significant factors for the occur- Page 3 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 1: Patients characteristics (n = 34) Characteristics Number of patients (%) Gender Male 25 (74) Female 9 (26) Age(yrs) Median 63 Range 42~78 ECOG 0 13 (38) 1 16 (47) 2 5 (15) Site Upper 22 (65) Lower 10 (29) Upper+Lower 2 (6) Histology Sqamous cell ca 14 (41) Adeno ca 6 (18) Small cell ca 11 (32) Other 3 (9) Location Central 26 (76) Peripheral 8 (24) Previous surgery No 22 (65) Yes 12 (35) Previous chemotherapy No 13 (38) Yes 21 (62) Concurrent chemotherapy No 19 (56) Yes 15 (44) Smoking Never 7 (21) Previous 13 (39 Current 13 (39) PFTs FEV1(L) Median 1.8 Range 0.8~4.5 FEV1/FVC(%) Median 67 Range 34~94 DLCO(%) Median 78 Range 53~133 Abbreviations: ECOG, Eastern Cooperative Oncology Group Performance Scale; PFTs, Pulmonary Function Tests rence of RP (p = 0.035, p = 0.049, and p = 0.049, respec- 0.062). However, TGF-β1 level of patients who did not tively). The V30 and MLD were marginal significant (p = develop RP began to decrease relative to their pretreat- 0.068, p = 0.077). For the lung as a separate organ, the ment level after middle of RT. MLD, V5, V10 and V20 values were statistically significant factors for the occurrence of RP (p = 0.018, p = 0.003, p = The pretreatment TGF-β1 level of the patient who died of pg/ml). The TGF-β1 level 0.006 and p = 0.032, respectively). The V30 was margin- RP was relatively high (2.9 × 10 ally significant (p = 0.053). decreased during RT as same as other RP patients. His TGF-β1 level at 4 weeks after RT was markedly increased Serum cytokine levels upto 4.2 × 10 pg/ml. TGF- 1 As demonstrated in Table 3, the mean pretreatment TGF- We performed an ANOVA model for repeated-measures β1 level was 2.8 ± 0.8 × 10 pg/ml for the RP group and for analysis of chronological change in TGF-β1 level and 2.3 ± 1.1 × 10 pg/ml for the non-RP group. The patients found that there were significant associations between the who developed pneumonitis showed a higher level of pre- changes of TGF-β1 level during the time course of radia- treatment TGF-β1, but this was not statistically significant tion and the risk of developing RP (p < 0.001 for develop- (p = 0.157). During the period of radiation treatment, ment of any grade of RP, p < 0.0001 for development of from the beginning of RT to the middle of RT, the RP grade ≥2 RP). These chronological changes in the serial group tended to show a decrease in the TGF-β1 level. TGF-β1 levels are demonstrated in Table 4, Fig 1. However, the TGF-β1 level began to increase at the end of RT in the RP group and became significantly higher at 4 IL-6 weeks after RT (p = 0.007). These elevation of TGF-β1 level The pretreatment IL-6 level was higher in the non-RP after RT were same in the patients of ≥ grade 2 RP (p = group compared to the RP group (49.1 ± 123.3 pg/ml vs Page 4 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 2: Dosimetric risk factors for development of RP ≥ grade 2 Mean ± SD DVH parameters RP Non-RP p value* Whole lung MLD (cGy) 1188 ± 354 957 ± 382 0.077 V5 (%) 58 ± 15 45 ± 12 0.035 V10 (%) 43 ± 13 34 ± 13 0.049 V20 (%) 29 ± 8 22 ± 11 0.049 V30 (%) 22 ± 7 17 ± 9 0.068 V40 (%) 15 ± 5 12 ± 7 0.299 Ipsilateral lung MLD (cGy) 1986 ± 491 1512 ± 612 0.018 V5 (%) 78 ± 13 63 ± 17 0.003 V10 (%) 67 ± 11 52 ± 19 0.006 V20 (%) 49 ± 11 38 ± 17 0.032 V30 (%) 38 ± 11 29 ± 15 0.053 V40 (%) 24 ± 9 22 ± 14 0.655 Abbreviations: RP, radiation pneumonitis; DVH, dose volume histogram; MLD, mean lung dose; V5, the percentage of the irradiated lung volume receiving a radiation dose exceeding 5 Gy; V10, the percentage of the irradiated lung volume receiving a radiation dose exceeding 10 Gy; V20, the percentage of the irradiated lung volume receiving a radiation dose exceeding 20 Gy; V30, the percentage of the irradiated lung volume receiving a radiation dose exceeding 30 Gy; V40, the percentage of the irradiated lung volume receiving a radiation dose exceeding 40 Gy * student's t-test 16.7 ± 23.9 pg/ml, respectively, p = 0.387). However, the Other cytokines IL-6 level between each patient showed wide variation The IL-1α, IL-10 and TNF-α level were stable during the compared to the TGF-β1 level, which showed a relatively whole course of RT and they were not correlated with the stable range of variation. The IL-6 level was low before risk of RP (data not shown). and during RT, but it began to increase after RT in both groups. The changes of the IL-6 during the time course Discussion were similar whether patients developed RP or not. Our Being aware of the risk of radiation pulmonary injury is an data showed a wide range of variation in the circulatory important aspect of patient management in the era of IL-6 levels, but relatively little changes with the develop- combination of chemoradiation for treating lung cancer. ment of RP. The recent studies on the new effective chemotherapeutic agents or hyperfractionated RT have reported that the inci- Table 3: Changes of mean TGF-β1 level during the course of dence of grade ≥ 3 RP in patients who underwent concur- radiation therapy rent chemoradiotherapy is between 21-23%[6,7]. Along Mean ± SD (× 10 pg/ml) with combined chemotherapy, other patient-specific or Time RP Non-RP p value* treatment-specific factors have been identified as predic- tors of developing RP. However, currently, there are no RP1 Before RT 2.8 ± 0.8 2.3 ± 1.1 0.157 generally accepted methods available to accurately predict (n = 17) Beginning of RT 2.4 ± 0.7 2.5 ± 1.1 0.738 an individual patient's risk of developing RT-induced pul- Middle of RT 2.0 ± 0.6 2.4 ± 1.2 0.19 monary morbidity. The purpose of this study was to assess End of RT 2.1 ± 0.7 2.2 ± 0.9 0.879 the values of the cytokine dynamics and dosimetric factors 2 wks after RT 2.1 ± 0.4 2.0 ± 0.8 0.806 to predict the risk of developing symptomatic RP. 4 wks after RT 2.8 ± 1.1 1.7 ± 0.5 0.007 RP2 Before RT 2.5 ± 1.0 2.6 ± 0.9 0.844 The dose of irradiation administered to patients is distrib- (n = 8) Beginning of RT 1.9 ± 0.6 2.6 ± 0.9 0.079 uted in a 3-D volume on the DVH. According to a litera- Middle of RT 1.8 ± 0.7 2.3 ± 0.9 0.168 ture review, dosimetric factors such as V5, V10, V20, V30, End of RT 1.9 ± 0.6 2.2 ± 0.8 0.352 V (effective volume) and MLD have statistically signifi- eff 2 wks after RT 2.1 ± 0.7 2.0 ± 0.9 0.787 cant correlation to symptomatic radiation pneumonitis 4 wks after RT 3.3 ± 1.7 2.0 ± 0.9 0.062 [9-11]. Also in the current study, most of the dosimetric factors showed an association with RP. Because the lung Abbreviations: RT, radiation therapy; wks, weeks; RP1, development of any grade of radiation pneumonitis; RP2, development of grade ≥ 2 function may not be uniform across all regions of the radiation pneumonitis; lung, it is unlikely that such a simplistic dose-volume rela- * student's t-test Page 5 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Table 4: Results of repeated measures ANOVA about the changes of mean TGF-β1 level during the course of radiation therapy Mean ± SD (× 10 pg/ml) Repeated measures ANOVA Before RT Beginning of Middle of RT End of RT 2 wks after RT 4 wks after RT Source F p value RT RP1 2.8 ± 0.8 2.4 ± 0.7 2.0 ± 0.6 2.1 ± 0.7 2.1 ± 0.4 2.8 ± 1.1 RP1 0.28 0.598 (n = 17) Non-RP 2.3 ± 1.1 2.5 ± 1.1 2.4 ± 1.2 2.2 ± 0.9 2.0 ± 0.8 1.7 ± 0.5 time 2.36 0.043 (n = 17) time*RP1 4.79 0.001 RP2 2.5 ± 1.0 1.9 ± 0.6 1.8 ± 0.7 1.9 ± 0.6 2.1 ± 0.7 3.3 ± 1.7 RP2 0.05 0.828 (n = 8) Non-RP 2.6 ± 0.9 2.6 ± 0.9 2.3 ± 0.9 2.2 ± 0.8 2.0 ± 0.9 2.0 ± 0.9 time 3.24 0.008 (n = 26) time*RP2 6.27 0.0001 Abbreviations: RT, radiation therapy; wks, weeks; RP1, development of any grade of radiation pneumonitis; RP2, development of grade ≥ 2 radiation pneumonitis tionship exists. However, these parameters are easy to cal- oped symptomatic pneumonitis [14]. Kong et al. reported culate and useful in the clinical setting. that TGF-β1 decreased during RT in patients with an increased pretreatment plasma TGF-β1 level, yet this Although dosimetric factors are important, these factors didn't normalize even by the completion of treat- don't take into consideration the molecular biological ment[19]. Other clinical studies have reported that the events that may be responsible for radiation-induced absolute level or the relative ratio of the TGF-β1 level responses of normal tissue. Molecular events have been showed meaningful changes during or after RT in patients shown to occur much earlier than the clinically apparent who suffered RP. However, the correlated time points radiation responses. Exposure to ionizing radiation rap- were different in each study and the patterns of the idly triggers a cascade of genetic and molecular events, changes were not always same [14,18-20]. We also evalu- which is an active process involving the production of a ated the TGF-β1 ratios (the ratios of a value from a partic- number of inflammatory and fibrogenic cytokines by var- ular time-point divided by the pre-RT value) presented in ious cells in lung, for example, macrophages, epithelial table 5. In the patients who experienced grade ≥2 RP, the cells, endothelial cells, pneumocytes and fibroblasts. TGF-β1 ratios tended to show a decrease than that of the These molecular processes are perpetuated beyond the non-RP group, from the beginning of RT to the end of RT. time point at which the acute insult has been removed. However, the TGF-β1 level began to increase at the 2 Several recent studies have shown that cytokines (IL-1, IL- weeks after RT in the RP group and became more higher 6, IL-8, IL-10, TNF-α, platelet-derived growth factor and at 4 weeks after RT (1.6 ± 1.0 vs 0.8 ± 0.3, p = 0.081). The TGF-β), surfactant apoproteins and cell adhesion mole- time point predictive of RP is different from the data of cules (ICAM-1, E-selectin) have important roles in RT- Zhao et al which was 4 weeks during 6 weeks' course of RT induced pulmonary injury [8,9,12-18]. A study of the [20]. A recent report from Zhao et al suggests the combi- dynamics of the serum cytokines during the early course nation of TGF-β1 and MLD may help stratify the patients of the treatment would be useful for predicting the risk of for their risk of RP to improve the predictive power [21]. pulmonary injury and for the early intervention of pneu- Their data show the incidence of RP was 4.3% in patients monitis. Because these cytokines are thought to be key with a TGF-β1 ratio ≤ 1 and MLD ≤ 20 Gy, and 66.7% in mediators of lung toxicity, many of them have been exam- those with a TGF-β1 ratio ≥ 1 and MLD ≥ 20 Gy. ined as potential early markers for radiation pneumonitis. The TGF-β1 levels in bronchoalveolar lavage(BAL) fluid Profibrogenic cytokine TGF-β1 is the most extensively from the irradiated area increased continuously during investigated among the various biological markers in radi- and after RT compared to the pretreatment levels in the RP ation-induced injuries. We observed that TGF-β1 group[17]. On the other hand, in several studies the pat- decreased during RT and it began to increase at the end of tern of the changes of the TGF-β1 level was not distinct RT in the patients who developed RP. There have been between the RP and non-RP groups [12,15]. The reasons similar reported results for the TGF-β1 dynamics in rela- for such conflicting results may be explained by the fact tion to the development of pneumonitis after radiother- that numerous factors can falsely increase TGF-β1 levels apy for lung cancer. Hur et al. found that TGF-β1 was and confound their predictive value for RP occurrence. decreased during RT and it was markedly increased at 2-4 First, the tumor stroma may be responsible for the pro- weeks after the completion of RT for patients who devel- duction of TGF-β1 in lung cancer patients. The mean TGF- Page 6 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 [15,17]. In this study, the IL-6 level was low before and during RT, but it began to increase after RT. These findings are similar those of Chen et al [24]. However, the changes of the IL-6 level were similar whether patients developed RP or not. This suggests that the IL-6 produced by the lung is not a major determinant of the circulating IL-6 levels. The mean IL-6 concentrations were significantly higher in the lung cancer patients than in the normal controls, and the patients with metastatic tumor had higher IL-6 levels than those patients with undisseminated disease, suggest- ing that neoplastic cells may produce IL-6 [25]. IL-6 is an acute phase inflammatory cytokine, and this would sug- gest that the measurement of circulating IL-6 can reflect the inflammatory state of the lung. The IL-6 levels fre- quently increase in patients suffering with several pulmo- nary diseases, including infectious pneumonia, interstitial pneumonia and chronic obstructive pulmonary dis- Th RP g Figure 1 e p rou attern of the ch p and the non-RP gr anges of the mean TGF- oup β1 level in the ease[17]. The pattern of the changes of the mean TGF-β1 level in the RP group and the non-RP group. The TGF-β1 There are other potential biological predictors of RP, but level began to increase at the end of RT in the RP group and none has been conclusively demonstrated to identify the it became significantly higher at 4 weeks after RT (p = 0.007). patients who are at a high risk of radiation-induced pul- However TGF-β1 level of patients who did not develop RP monary toxicity [9,12,15,16]. Host-associated diseases, began to decrease relative to their pretreatment level after the proportion of gross disease at the time of irradiation middle of RT. The solid line shows the mean level of TGF-β1 and pre-RT treatment such as chemotherapy are all associ- in the RP group and the dashed line shows the mean level of TGF-β1 in the non-RP group. The data are presented as ated with local cytokine production [5,15]. The addition mean ± standard error of the mean. of radiation on a background of subclinical damage may augment a cytokine cascade and increase the severity of acute and late side effects. β1 level in lung cancer patients was higher than that in the normal controls (p < 0.001). According to a pathology Although various cytokines are important in the patho- slide review, the degree of fibrosis that is present in tumor genesis of radiation-induced pulmonary injury, our is also significantly correlated with an elevated plasma results show that the changes of TGF-β1 could be an inde- TGF-β1 level (p = 0.03). The dynamics of the plasma TGF- pendent predictor of developing RP and well correlated β1 have been suggested to be a marker of RT-induced nor- with the time course of radiation therapy. However, the mal tissue injury as well as a marker of tumor response. role of cytokine markers in the cytokine cascades that pro- After radiotherapy, the patients who were alive with dis- mote pulmonary injury deserves further investigation. ease had significantly higher TGF-β1 levels than those Also, treatment strategies designed to block this patho- who were alive with no evidence of disease (p = 0.02) [19] logic process may need to be continued well beyond the Second, careful handling of the sample is also important. completion of RT. Variations of the centrifugation conditions and platelet contamination could artificially elevate plasma TGF-β1 Table 5: Changes of mean TGF-β1 ratio during the course of level[19,22] radiation therapy Mean of Ratio ± SD IL-6 is a pleiotropic inflammatory cytokine that is impor- Time RP Non-RP p value* tant in regulating immunologic and inflammatory responses. IL-6 levels before, during and after thoracic RT RP2 Before RT were significantly higher in those patients who developed (n = 8) Beginning of RT 0.8 ± 0.2 1.0 ± 0.2 0.019 pneumonitis in several reports [13,23]. Especially Arpin et Middle of RT 0.8 ± 0.4 0.9 ± 0.4 0.238 al reported covariation of proinflammatory cytokine (IL- End of RT 0.8 ± 0.2 0.9 ± 0.3 0.346 2 wks after RT 1.0 ± 0.6 0.9 ± 0.3 0.576 6) and anti-inflammatory cytokine (IL-10) levels during 4 wks after RT 1.6 ± 1.0 0.8 ± 0.3 0.081 the first 2 week of RT were independent predictive evi- dence of RP. However, other studies have failed to find a Abbreviations: RT, radiation therapy; wks, weeks; RP2, development relationship between IL-6 and the radiation-induced pul- of grade ≥ 2 radiation pneumonitis monary symptoms, like what our results have shown * student's t-test Page 7 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 small-cell lung cancer: radiation therapy oncology group Conclusion protocol 91-06. J Clin Oncol 1996, 14:1055-1064. Although the current study had a limited number of 7. Kouroussis C, Mavroudis D, Kakolyris S, Voloudaki A, Kalbakis K, patients, we demonstrated that, in the patients who devel- Souglakos J, Agelaki S, Malas K, Bozionelou V, Georgoulias V: High incidence of pulmonary toxicity of weekly docetaxel and oped RP, the TGF-β1 level decreased during RT and began gemcitabine in patients with non-small cell lung cancer: to increase at the end of RT. And the TGF-β1 level at 4 results of a dose-finding study. Lung cancer 2004, 44:363-368. 8. Claude L, Perol D, Ginestet C, Falcheroc L, Arpind D, Vincente M, weeks after RT was significantly higher than the TGF-β1 Martela I, Hominalf S, Cordierg JF, Carrie C: A prospective study level of the patients who didn't develop RP. TGF-β1 may on radiation pneumonitis following conformal radiation contribute to the process leading to a radiation injury in therapy in non-small-cell lung cancer: clinical and dosimetric factors analysis. Radiother Oncol 2004, 71:175-181. human lung tissue. While the change of TGF-β1 level did 9. Wang S, Liao Z, Wei X, Liu HH, Tucker SL, Hu CS, Mohan R, Cox JD, not take place early in time course of RT with having a pre- Komaki R: Analysis of clinical and dosimetric factors associ- dictive value in our study, the incorporation of the biolog- ated with treatment-related pneumonitis (TRP) in patients with non-small-cell lung cancer(NSCLC) treated with con- ical parameters into the dosimetric data that have been current chemotherapy and three-dimensional conformal developed to predict radiation-induced lung injury may radiotherapy(3D-CRT). Int J Radiat Oncol Biol Phys 2006, 66:1399-1407. improve the predictive accuracy. Further research must 10. Vujaskovic Z, Marks LB, Anscher MS: The physical parameters continue to identify biomarkers that will one day allow us and molecular events associated with radiation-induced lung to tailor our therapies in response to the highly accurate toxicity. Semin Radiat Oncol 2000, 10:296-307. 11. Hartsell WF, Scott CB, Dundas GS, Mohiuddin M, Meredith RF, Rubin predictions of risk for the development of radiation pneu- P, Weigensberg IJ: Can serum markers be used to predict acute monitis. and late toxicity in patients with lung cancer ? analysis of RTOG 91-03. Am J Clin Oncol 2007, 30:368-376. 12. Chen Y, Williams J, Ding I, Hernady E, Liu W, Smudzin T, Finkelstein Competing interests JN, Rubin P, Okunieff P: Radiation pneumonitis and early circu- The authors declare that they have no competing interests. latory cytokine markers. Semin Radiat Oncol 2002, 12(Suppl 1):26-33. 13. Chen Y, Rubin P, Williams J, Hernady E, Smudzin T, Okunieff P: Cir- Authors' contributions culating IL-6 as a predictor of radiation pneumonitis. Int J JYK performed the collection of blood samples, acquisi- Radiat Oncol Biol Phys 2001, 49:641-648. 14. Hur WJ, Yoon SM, Lee HS, Yang KM, Shin GH, Son CH, Han JY, Lee tion of clinical data and drafted the manuscript. YSK KN, Jeong MH: The measurements of plasma cytokines in designed and coordinated the study, checked statistical radiation-induced pneumonitis in lung cancer patients. J Korean Soc Ther Oncol 2000, 18:314-320. results, read and edited the manuscript. YKK coordinated 15. Hart JP, Broadwater G, Rabbani Z, Moeller BJ, Clough R, Huang D, and performed laboratory work. HJP interpreted radiolog- Sempowsk GA, Dewhirst M, Pizzo SV, Vujaskovic Z, Anscher MS: ical findings. SJK, JHK, YPW, SCY, SNL and HSJ performed Cytokine profiling for prediction of symptomatic radiation- induced lung injury. Int J Radiat Oncol Biol Phys 2005, 63:1448-1454. evaluation of patients and read the manuscript. All the 16. Ishii Y, Kimura S: Soluble intercellular adhesion molecule-1 as authors read and approved the final manuscript. an early detection marker for radiation pneumonitis. Eur Respir J 1999, 13:733-138. 17. Barthelemy-Brichant N, Bosquee L, Cataldo D, Corhay JL, Gustin M, Acknowledgements Seidel L, Thiry A, Ghaye B, Nizet M, Albert A, Deneufbourg JM, Bar- The authors wish to acknowledge the financial support of the Catholic tsch P, Nusgens B: Increased IL-6 and TGF-β1 concentrations Medical Center Research Foundation made in the program year of 2006. in bronchoalveolar lavage fluid associated with thoracic radi- otherapy. Int J Radiat Oncol Biol Phys 2004, 58:758-767. 18. Novakova-Jiresova A, Gameren MM, Coppes RP, Kampinga HH, References Groen H: Transforming growth factor-β plasma dynamics and 1. Willner J, Caragiani E, Tschammler A, Flentje M: Dose, volume, and post-irradiation lung injury in lung cancer patients. Radiother tumor control prediction in primary radiotherapy of non- Oncol 2004, 71:183-189. small-cell lung cancer. Int J Radiat Oncol Biol Phys 2002, 19. Kong FM, Washington MK, Jirtle RL, Anscher MS: Plasma trans- 52:382-389. forming growth factor-β1 reflects disease status in patients 2. Kong FM, Ten Haken RK, Schipper MJ, Sullivan MA, Chen M, Lopez with lung cancer after radiotherapy: a possible tumor C, Kalemkerian GP, Hayman JA: High-dose radiation improved marker. Lung Cancer 1996, 16:47-59. local tumor control and overall survival in patients with inop- 20. Zhao L, Sheldon K, Chen M, Yin MS, Hayman JA, Kalemkerian GP, erable/unresectable non-small-cell lung cancer: Long-term Arenberg D, Lyons SE, Curtis JL, Davis M, Cease KB, Brenner D, results of a radiation dose escalation study. Int J Radiat Oncol Anscher MS, Lawrence TS, Kong FM: The predictive role of Biol Phys 2005, 63:324-333. plasma TGF-beta1 during radiation therapy for radiation- 3. Marino P, Preatoni A, Cantoni A: Randomized trials of radiother- induced lung toxicity deserves further study in patients with apy alone versus combined chemotherapy and radiotherapy non-small cell lung cancer. Lung Cancer 2008, 59:232-239. in stages IIIa and IIIb non-small cell lung cancer a meta-anal- 21. Zhao L, Wang L, Ji W, Wang X, Zhu X, Hayman JA, Kalemkerian GP, ysis. Cancer 1995, 76:593-601. Yang W, Brenner D, Lawrence TS, Kong FM: Elevation of plasma 4. Chevalier TL, Arriagada R, Quoix E, Ruffie P, Martin M, Douillar JY, TGF-beta1 during radiation therapy predicts radiation- Tarayre M, Lacombe-Terrier MJ, Laplanche A: Radiotherapy alone induced lung toxicity in patients with non-small-cell lung can- versus combined chemotherapy and radiotherapy in unre- cer: a combined analysis from Beijing and Michigan. Int J sectable non-small cell lung carcinoma. Lung Cancer 1994, Radiat Oncol Biol Phys 2009, 74:1385-1390. 10(Suppl 1):S239-S244. 22. Wakefield LM, Letterio J, Chen J, Danielpour D, Allison RS, Pai LH, 5. Mehta V: Radiation pneumonitis and pulmonary fibrosis in Denicoff AM, Noone MH, Cowan KH, O'Shaughnessy JA: Trans- non-small-cell lung cancer: Pulmonary function, prediction, forming growth factor-beta1 circulates in normal human and prevention. Int J Radiat Oncol Biol Phys 2005, 63:5-24. plasma and is unchanged in advanced metastatic breast can- 6. Lee JS, Komaki R, Fossella FV, Dundas GS, McDonald S, Byhardt RW, cer. Clin Cancer Res 1995, 1:129-136. Curran WJ Jr: Concurrent chemoradiation therapy with oral 23. Arpin D, Perol D, Blay JY, Falchero L, Claude L, Vuillermoz-Blas S, etoposide and cisplatin for locally advanced inoperable non- Martel-Lafay I, Ginestet C, Alberti L, Nosov D, Etienne-Mastroianni B, Page 8 of 9 (page number not for citation purposes) Radiation Oncology 2009, 4:59 http://www.ro-journal.com/content/4/1/59 Cottin V, Perol M, Guerin JC, Cordier JF, Carrie C: Early variations of circulating interleukin-6 and interleukin-10 levels during thoracic radiotherapy are predictive for radiation pneumo- nitis. J Clin Oncol 2005, 23:8748-8756. 24. Chen Y, Hyrien O, William J, Okunieff P, Smudzin T, Rubin P: Inter- lekin (IL)-1A and IL-6: Application to the predictive diagnos- tic testing of radiation pneumonitis. Int J Radiat Oncol Biol Phys 2005, 62:260-266. 25. De Vita F, Orditura M, Auriemma A, Infusino S, Roscigno A, Catalano G: Serum levels of interleukin-6 as a prognostic factor in advanced non-small cell cancer. Oncology reports 1998, 5:649-652. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 9 of 9 (page number not for citation purposes)

Journal

Radiation OncologySpringer Journals

Published: Nov 27, 2009

References