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

Learn More →

Analysis of PD-1, PD-L1, and T-cell infiltration in angiosarcoma pathogenetic subgroups

Analysis of PD-1, PD-L1, and T-cell infiltration in angiosarcoma pathogenetic subgroups Angiosarcoma (AS) is a rare malignancy with a poor prognosis. It can develop spontaneously or due to previous radiotherapy (RT), ultraviolet (UV) radiation, or lymphoedema (Stewart Treves AS). Novel therapeutic approaches are needed, but progress is hindered because of the heterogeneity and rarity of AS. In order to explore the potential of immune checkpoint inhibition (ICI), we investigated the protein expression of programmed cell death 1 (PD-1), programmed death-ligand 1 (PD-L1), and CD8 + T cells in 165 AS cases in relation to AS subgroups based on clinical classification and in relation to whole-genome methylation profiling based clusters (A1, A2, B1, B2). High PD-L1 and PD-1 expression were predominantly shown in UV-associated, visceral, and soft tissue AS. RT-associated AS showed predominantly high PD-1 expression. CD8 + T cell infiltration w as present in the majority of AS samples. Within the UV-associated AS, two different clusters can be distinguished by DNA methylation prol fi ing. Cases in cluster A1 showed higher PD-1 (p = 0.015), PD-L1 (p = 0.015), and CD8 + T cells (p = 0.008) compared to those in cluster B2, suggesting that these UV-AS tumors are more immunogenic than B2 tumors showing a difference even within one subgroup. In soft tissue AS, combined PD-1 and PD-L1 expression showed a trend toward poor survival (p = 0.051), whereas in UV-associated AS, PD-1 expression correlated with better survival (p = 0.035). In conclusion, we show the presence of PD-1, PD-L1, and CD8 + T cells in the majority of AS but reveal differ - ences between and within AS subgroups, providing prognostic information and indicating to be predictive for ICI. Keywords Angiosarcoma · Subgroups · Programmed cell death 1 · Programmed death-ligand 1 (PD-L1) · Immune checkpoint inhibition Introduction which DNA damaging factors including radiation, UV light exposure or chronic lymphoedema play an important role AS is a rare and aggressive vasoformative sarcoma aris- [1–3]. Current treatment options include surgery, RT, and/ ing at different anatomical sites, including skin, soft tissue, or chemotherapy, depending on the extent of the disease. In bone, and visceral organs. AS can be clinically classified into addition, the multi-tyrosine kinase inhibitor pazopanib is also primary AS (with unknown etiology) or secondary AS, in applied in daily practice [4] and treatment with the generic ß-blocker propranolol has been suggested [5]. The survival of AS patients is poor with a reported 5-year survival of *Among others; Jos W.R. Meyer, Rijnstate Hospital, Arnhem, The only 30–40% [6–8], emphasizing the need for novel treat- Netherlands and Marieke C.H. Hogenes, LABPON, Hengelo, The Netherlands. ment options. A potential approach for AS treatment is ICI. Tumor cells * Yvonne M. H. Versleijen-Jonkers can upregulate PD-L1 on their membrane to promote immune yvonne.versleijen-jonkers@radboudumc.nl suppression. Interaction with the receptor programmed cell Department of Pathology, Radboud University Medical death 1 (PD-1) on CD8 + T cells renders the T cell inactive, Center, Nijmegen, The Netherlands and thus prevents the killing of tumor cells. ICI with anti- Department of Medical Oncology (Internal Postal Code: PD-1 antibodies can reactivate the cytotoxic function of T 452), Radboud University Medical Center, P.O. Box 9101, cells leading to the subsequent killing of tumor cells [9]. 6500 HB Nijmegen, The Netherlands In order to determine the role of ICI in AS, it is necessary Princess Máxima Center for Pediatric Oncology, Utrecht, to examine the expression of PD-L1, PD-1, and the presence The Netherlands 1 3 Immunologic Research (2022) 70:256–268 257 of CD8 + T cells as potential biomarkers in this respect. Sev- pathogenetic subgroups based on available clinical data and eral studies investigated the expression of these biomarkers pathology reports [35]. in AS, reporting variable levels of expression and varying Tumor samples were collected on tissue microarrays correlations with prognosis (shown in Table  1). Of note, (TMAs) and divided into the different subgroups, includ - most studies were performed on cutaneous (predominantly ing 44 UV-associated, 14 cutaneous not UV-associated, 55 UV-associated) AS or small numbers of other subtypes, RT-associated, 14 Stewart Treves (lymphoedema-associated often not further specified. cases), 27 visceral, and 11 soft tissue cases. Only primary Clinical data with regard to the application of ICI in AS is localized tumor samples were selected for this study. Of limited to small case series. In one of these case series, three these samples, 33 have been previously subject to DNA UV-associated AS were treated with anti-PD-1 (one in com- methylation profiling [ 33]. bination with CTLA-4 inhibition) showing partial response. One RT-associated AS also showed partial response (with Clinical data anti-PD1) and one primary breast AS showed progressive disease on axitinib (VEGFR inhibitor) combined with PD-1 Clinical data were received from the nationwide Netherlands inhibition [27]. Four patients suffering from UV-associated Cancer Registry and were linked to data from the Dutch AS with significant PD-L1 expression were successfully pathology registry (PALGA). Ethical approval for the study treated with anti-PD-1 [28–31]. In the recent AS patient- was obtained from the local certified Medical Ethics Com - partnered-genomic study of Painter et  al., a total of six mittee of the Radboudumc, Nijmegen, The Netherlands (file patients were treated with anti-PD-1 [32]. Of these patients, number 2016–2686). two out of three UV-associated AS with high mutational burden (> 150 mutations/Mb) showed a complete response, Immunohistochemistry whereas no clinical benefit was found in the other three patients (all non-UV AS with low mutational burden) [32]. Immunohistochemical analysis was performed to inves- Pecora et al. published a case of primary AS of the temple tigate PD-1 and PD-L1 expression and the presence of and one RT-associated AS with complete clinical remission CD8 + lymphocytes in the tumor. Tonsil (PD-L1 + , PD-1 + , on combined anti-CTLA4 and anti-PD-1 therapy (Pecora and CD8 +) and appendix (CD8 + and PD1 +) served as et al. CTOS2019). positive controls. Immunohistochemistry was performed Taken together, the previously mentioned studies sug- on 4-µm-thick FFPE sections of AS TMAs with one or two gest a potential therapeutic role for ICI in especially UV- cores per sample from representative tumor areas (core size associated AS. However, it yet remains unclear whether 2 mm) to allow simultaneous examination of patient speci- immunotherapy could be of interest for all AS patients or mens under identical conditions. Staining was performed in only for certain pathogenetic subgroups. Our recent DNA the Lab Vision Autostainer 360 (Thermo Fisher Scientific) methylation profiling study confirmed the existence of these by using the EnVision FLEX, pH high Link kit (Dako), and subgroups of AS on an epigenetic level, which did not fully monoclonal rabbit anti-PD-L1 (1:800, clone E1L3N, Cell match the clinical subtypes [33]. Signaling Technology), monoclonal mouse anti-PD-1 (1:20, In the current retrospective study, we aimed to character- clone MRQ-22, Cell Marque) or monoclonal mouse anti- ize PD-1, PD-L1 expression, and the presence of CD8 + T CD8 (1:80, clone C8/144B, Dako). cells in a large cohort of AS tumor samples and their prog- PD-L1 expression on the tumor cells was scored as 0% nostic relevance to further explore the heterogeneity and the ( −), 1–10% (+ / −), 10–50% ( +) or ≥ 50% positive tumor need to differentiate between the different AS subgroups. cells (+ +). All CD8 and PD-1 positive T cells were counted and subdivided in three categories: < 10 ( −), 10–50 ( +), or ≥ 50 positive cells (+ +) per tumor core [36]. Digital images were generated with VisionTekTM Materials and methods (Sakura, version 2.6) and analyzed at × 20 magnification. Tumor sample collection Statistical analysis We collected formalin-fixed paraffin-embedded (FFPE) Statistical analyses were performed using IBM SPSS Sta- tumor tissue of AS patients by a nationwide search through tistics 25. p-values < 0.05 were considered significant and PALGA (Dutch nationwide network and registry of histo- p-values < 0.1 were considered a trend. Relations between and cytopathology) diagnosed between 1989 and 2015 in categorical parameters were assessed by chi-square or Fish- the Netherlands [34]. All cases were reviewed by an expert er’s exact testing as appropriate, and associations with over- pathologist (UF), and confirmed AS cases were divided into all survival (OS) were assessed by the Kaplan–Meier method 1 3 258 Immunologic Research (2022) 70:256–268 1 3 Table 1 Overview of studies regarding the presence of PD-1, PD-L1, and CD8 in AS Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Kawamura 29 Cutaneous 12 (41%) 22 (76%) 10 (34%) > 5% Mouse anti- Rabbit anti- PD-L1 et al. 2019 human ab human ab expression [10] (CST) (CST) associated with poor survival Gambichler 12 Cutaneous 4 (33%) 5 (42%) Ab92484 Ab205921 et al. 2020 (Abcam) (Abcam) [11] Fujii et al. 55 Cutaneous 20/40 (50%) ≥ Median/ C8/144B High number 2014 [12] (mainly HPF (Dako) of CD8 in UV) primary tumor cor- related with improved survival Honda et al. 106 Cutaneous 19 (18%) 32 (30%) 9 (9%) > 50/3HPF > 5% NAT105 SP142 PD-1 + cells 2017 [13] (mainly (Abcam) (Spring (especially UV) Biosience) in combina- tion with PD-L1 expression) corre- lated with improved survival Shimizu et 52 Cutaneous 21 (40%) 24 (46%) ≥ 5% > Median Rabbit mon- Abcam PD-L1 and al. 2017 (mainly oclonal CD8 are [14] UV) antibody associ- (Abcam) ated with a worse outcome 3 3 2 Bagaria et 26 Cutaneous 1 (4%) 5 (19%) ≥ 5% NAT105 SP-142 and No al. 2018 (19), other (Ventana) SP-263 [15] (8) (Ventana) Immunologic Research (2022) 70:256–268 259 1 3 Table 1 (continued) Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Botti et al. 24 Primary AS 16 (66%) ≥ 5% SP-142 No 2017 [16] (7 breast, (Spring 5 soft Biosience) tissue, 4 bone, 4 skin, 4 visceral) Googe et al. 10 Skin (9), 10 (100%) 10 (100%) 10 (100%) 10 (100%) Low: ≥ 1% NAT105 ZR 3 4B11 (Leica 2020 [17] soft tissue High: ≥ 50% (Cell (Cell Biosys- (1) Marque) Marque) tems) D’Angelo et 3 n.s 0 (0%) 1 (33%) ≥ 1% ≥ 5% Dako C8/144B al. 2015 (Dako) [18] Kim et al. 5 n.s 4 (80%) 4 (80%) 4 (80%) ≥ 1 Total NAT H-130 2013 [19] score ≥ 8 (Abcam) (SCB) Boxberg et 23 n.s 8 (35%) ≥ 1% SP263 (Ven- al. 2018 4 (17%) ≥ 5% tana) [20] 3 (13%) ≥ 10% 2 (9%) ≥ 50% Kosemehme- 7 n.s 1 (14%) > 5% E1L3N toglu et al. (CST) 2017 [21] Que et al. 5 n.s 1 (20%) > 1% E1L3N 2017 [22] (CST) Blessin et al. 25 n.s Median TC8 2020 [23] count 95 (ONCOdi- cells/mm anova) Orth et al. 6 n.s 2/5 3 (50%) 1 (17%) ≥ 4/HPF > 1% 315 M (Cell E1L3N 2020 [24] (40%) Marque) (CST) Vargas et al. 17 n.s 5 (29%) ≥ 1% SP263 (Ven- 2020 [25] tana) Lee et al. 70 n.s 13 (19%) Combined 22C3 PD-L1 2021 [26] positive (Agilent expression score ≥ 1 Technolo- correlated gies) with poor survival in metastatic AS patients 260 Immunologic Research (2022) 70:256–268 1 3 Table 1 (continued) Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Summary of N AS subtype PD-1 PD-L1 PD-1 and CD8 PD-1 PD-L1 CD8 results expres- expres- PD-L1 expres- threshold threshold threshold sion (%) sion (%) expres- sion (%) sion (%) 286 Cutaneous 44/156 92/212 28/144 53/101 > 50/3HPF ≥ 1% ≥ Median (28%) (43%) (19%) (52%) to ≥ 5% 7 Breast 4/7 (57%) ≥ 5% 6 Soft tissue 1/1 (100%) 5/6 (83%) 1/1 (100%) 1/1 (100%) ≥ 1% to ≥ 5% 4 Bone 4/4 (100%) ≥ 5% 4 Visceral 1/4 (25%) ≥ 5% 169 n.s 6/10 (60%) 35/136 5/11 (45%) 1/3 (33%) ≥ 1 to ≥ 4/ ≥ 1% to ≥ 8 ≥ 5% (26%) HPF Abbreviations: AS, angiosarcoma; CST, Cell Signaling Technology, HPF, high power field; n.s., not specified; SCB, Santa Cruz Biotechnology; UV, UV-associated AS; For PD-L1, the staining intensity score was classified as 0 (no staining), 1 (weak staining), and 2 (intermediate staining), and 3 (strong staining). The area of staining was scored as 0 (0–10% of the cells stained), 1 (11–33% of the cells stained), 2 (34–66% of the cells stained), and 3 (67–100% of the cells stained). The total score was determined as the sum of the intensity score and 2 3 4 the staining proportion score of two different TMAs. The total score ≥ 8 was determined as positive. Membranous expression of tumor cells. Expression not displayed per subtype. Results are summarized per subtype Immunologic Research (2022) 70:256–268 261 with the logrank test. Tumors positive for one marker or between the different levels of expression was made in the a combination of markers were compared to tumors nega- analysis. tive for that particular marker or combination. A distinction Results Table 2 Patient characteristics Immune profiles in AS subtypes N (%) AS samples 165 The expression of PD-1, PD-L1, and CD8 was assessed in UV associated 44 (27) 165 AS samples divided over 6 different subgroups. Patient Cutaneous not UV associated 14 (8) characteristics are shown in Table  2. Staining results are RT associated 55 (33) presented in Table 3 with an example of each staining shown Stewart Treves 14 (8) in Supplementary Fig. 1. Visceral 27 (16) High PD-1 and PD-L1 expressions were predominantly Soft tissue 11 (7) seen in clinically defined soft tissue (40%), UV-associated Extent of disease (18%), and visceral (17%) AS subgroups. Besides, RT-   Localized 80 (48) associated AS showed predominantly high PD-1 expression   Invasion adjacent structures 9 (5) (32%). Infiltration of high numbers of CD8 + T cells was   Lymph node involvement 4 (2) present in the majority of AS samples across all different   Distant metastases 14 (8) subgroups (64–86%) (Table 3).   Unknown 58 (35) The two main clusters (A and B) defined by our previ - Tumor depth ous genome-wide array-based DNA methylation profiling   Superficial 44 (27) study were each subdivided into 2 separate clusters (A1,   Deep 5 (3) A2 and B1, B2) (Fig. 1A). Cluster A1 consisted exclusively   Unknown 116 (70) of UV-associated cases, whereas A2 primarily consisted of Distant metastases RT-associated cases. Cluster B1 had both visceral and soft   No 111 (67) tissue cases, and cluster B2 was mixed, including cases of   Yes 20 (12) UV-associated AS.   Unknown 34 (21) In the current study, a significantly higher PD-L1 expres - Age sion (≥ 10%) and PD-1 expression (≥ 10) was found in the    < 40 6 (4) UV-associated cases in cluster A1 versus those in cluster B2    ≥ 40 < 70 55 (33) (for both stainings 6/7 (86%) in A1 versus 0/4 (0%) in B2,    ≥ 70 104 (63) p = 0.015) (Fig. 1B). Gender A high amount of CD8-positive T cells (≥ 50) was   Male 54 (33) observed in all UV-associated cases in cluster A1 versus   Female 111 (67) none of the UV-associated cases in cluster B2 (7/7 (100%) Follow-up status in A1 versus 0/3 (0%) in B2 (1 case was not evaluable), (median follow-up 14.8 months) p = 0.008) (Fig. 1B). A difference in the number of copy   Alive 20 (12) number variations (CNVs) was seen between the UV-associ-   Deceased 145 (88) ated AS cases in cluster A1 versus B2, as already described Table 3 PD-L1, PD-1 expression, and the presence of CD8 + T cells in angiosarcoma AS subgroup N PD-L1 ≥ 1% PD-L1 ≥ 10% PD-L1 ≥ 50% PD-1 ≥ 10 PD1 ≥ 50 CD8 ≥ 10 CD8 ≥ 50 PD-1 and PD-1 and PD-L1 ≥ 10 PD-L1 ≥ 50 (%) (%) UV associated 44 89% 66% 37% 66% 39% 98% 79% 50% 18% Cutaneous not UV 14 86% 50% 7% 64% 7% 93% 64% 21% 0% RT associated 55 79% 52% 8% 68% 32% 98% 81% 39% 6% Stewart Treves 14 75% 50% 8% 43% 7% 100% 86% 36% 0% Visceral 27 79% 54% 46% 52% 20% 92% 69% 38% 17% Soft tissue 11 100% 78% 56% 72% 36% 91% 82% 60% 40% 1 3 262 Immunologic Research (2022) 70:256–268 Fig. 1 Overview of the division of AS subgroups over the dif- ferent methylation clusters (A). Expression of PD-1, PD-L1, CD8, and the presence of copy number variations (CNVs) in UV-associated AS in cluster A1 versus B2 (B) A1 A2 B1 B2 Cluster A1 Cluster B2 PD-1 14% <10 43% ≥10<50 100% ≥50 43% PD-L1 14% <10 43% 50% 50% ≥10<50 ≥50 43% CD8 <10 25% ≥10<50 ≥50 100% unknown 75% CNVs <10 25% ≥10<20 ≥20 75% 100% 1 3 Immunologic Research (2022) 70:256–268 263 in our methylation profiling study (mean number of CNVs (19%) female, p = 0.039). Also, combined positive PD-L1 34.4 (range 23–39) in cluster A1 versus 7.3 (range 3–17) in and CD8 (≥ 10(%)) expression and high PD-L1 and CD8 cluster B2, p < 0.001) (Fig. 1B). (≥ 50(%)) expression showed a trend toward a correlation For the other AS subgroups, no significant differences in with the male gender (p = 0.070/0.075). High PD-L1 expres- expression of PD-L1, PD-1, or CD8 were observed between sion (≥ 50%) was also more common in deep tumors (3/4 clusters (Supplementary Fig. 2). (75%) deep versus 6/40 (15%) superficial tumors, p = 0.023). In RT-associated AS, PD-L1 expression (≥ 1%) showed Prognostic relevance of the immune profile a trend toward a correlation with the absence of metastases (p = 0.066), whereas no significant correlation with the pres - In Supplementary Table 1, we present the univariate analysis ence of distant metastases was observed at the thresholds of associations of the expression of the different markers of 10 or 50%. High PD-L1 expression (≥ 50%) did show with overall survival. In the total group PD-1, PD-L1, CD8, a significant correlation with the age below 70 years (4/21 or combined expression did not significantly correlate with (19%) < 70  years of age versus 0/27 (0%) ≥ 70  years of survival, although we did observe a trend toward a worse age, p = 0.031). All cases with high PD-L1 expression also overall survival for patients with ≥ 10% PD-L1 expres- showed a high CD8 expression. The combination of both sion in their tumor versus < 10% PD-L1 (median 11.0 ± 2.4 high PD-L1 and PD-1 expressions, as well as the combina- versus 17.1 ± 4.1  months, respectively, p = 0.088) and for tion of high PD-L1, PD-1, and CD8 expressions showed a patients with both ≥ 10% PD-L1 expression and ≥ 10 CD8- trend toward a correlation with the age group below 70 years positive T cells in their tumor versus those with no expres- (p = 0.077). High PD-1 expression (≥ 50) correlated with sion of both markers (median 11.0 ± 2.4  months versus the presence of distant metastases (PD-1 ≥ 50 in 3/3 (100%) 18.4 ± 4.1 months, respectively, p = 0.083) (Fig. 2A). patients with distant metastases versus 11/27 (41%) without In the cutaneous not UV-associated, RT-associated, Stew- distant metastases, p = 0.034). Cases with high PD-1 expres- art Treves, and visceral AS groups, no significant correla - sion also showed high CD8 expression. tions with survival were observed. We excluded correlations In visceral AS, high PD-L1 expression (≥ 50%) corre- when groups of only one patient were involved. lated with the absence of distant metastases (PD-L1 ≥ 50% in In patients with soft tissue AS, the presence of both PD-1 0/5 (0%) patients with distant metastases versus 8/13 (62%) (≥ 10) and PD-L1 (≥ 10%) showed a trend toward poor sur- without distant metastases, p = 0.036). vival (estimate mean survival time 6.8 ± 2.7 months (both positive, n = 6) versus 145.5 ± 41.3 months (not both posi- tive, n = 4), p = 0.051) (Fig. 2B). Discussion In UV-associated AS, the presence of PD-1 positive cells in the tumor (≥ 10) correlated with better overall This is the first study mapping the immunological landscape survival (median 20.6 ± 8.5  months for PD1 ≥ 10 versus in different AS subgroups as well as in genome-wide meth - 5.8 ± 2.1 months for PD1 < 10, p = 0.035). The same applies ylation profiling clusters. We detected relevant differences to the presence of high numbers of PD-1 positive cells in between the various subgroups. the tumor (≥ 50) (median 22.1 ± 16.7 months for PD1 ≥ 50 We showed a high expression of both PD-1 and PD-L1 versus 8.3 ± 2.8 months for PD1 < 50, p = 0.020) (Fig. 2C). predominantly in UV-associated, visceral, and soft tissue All PD-1 positive tumors in this subgroup were also CD8 AS subgroups and high PD-1 expression in RT-associated positive. AS, whereas infiltration of CD8 + T cells was present in the UV-associated AS patients in cluster A1 (instable, more majority of AS samples. In soft tissue AS the presence of immunogenic (“hot”) cluster, n = 7) showed a trend toward both PD-1 and PD-L1 expression showed a trend toward better overall survival compared to those in cluster B2 poor survival, whereas in UV-associated AS, PD-1 expres- (stable, “cold” cluster, n = 4) (median 22.2 ± 11.4 versus sion was correlated with better survival. 1.6 ± 4.5 months, p = 0.053) (Fig. 2D). These results reflect the heterogeneity in immunological response associated with prognosis for the diverse AS sub- Correlations with patient characteristics groups and underline the need to differentiate between them. So far, most studies have analyzed PD-1, PD-L1, and CD8 Correlations with gender, age, tumor depth, and presence expression only in cutaneous, mainly UV-associated AS of distant metastases are presented in Table 4. In the total (displayed in Table 1). Based on the observed high expres- group, PD-L1 expression (≥ 10%) already showed a trend sion of PD-1, PD-L1, and CD8 in visceral and soft tissue AS toward a correlation with the male gender (p = 0.075), in the current study, ICI might also be successful in these whereas high PD-L1 expression (≥ 50%) correlated signifi - subgroups. Individual immune profiling before the start of cantly with the male gender (17/47 (36%) male versus 19/98 ICI could be considered to select more vulnerable tumors. 1 3 264 Immunologic Research (2022) 70:256–268 Fig. 2 Kaplan–Meier curves showing the significant differences and ence of CD8 + T cells in the tumor and a Kaplan–Meier curve show- trends in overall survival in A AS total group, B soft tissue AS, and C ing the difference in overall survival between UV-associated AS in UV-associated AS according to PD-1, PD-L1 expression, or the pres- cluster A1 versus B2 (D) 1 3 Immunologic Research (2022) 70:256–268 265 Table 4 Correlations between (combinations of) biomarkers and clinical data AS subgroup Clinical marker PD-L1 PD-1 CD8 PD-L1 and PD-L1 and PD-1 and PD-L1 PD-1 CD8 CD8 and PD-1 and CD8 1 10 50 10 50 10 50 10 50 10 50 10 50 10 50 All Gender - 0.075 0.039 - - - - - - 0.070 0.075 - - - - Age - - - - - - - - - - - - - - - Tumor depth - - 0.023 - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - UV associated Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - Cutaneous not UV Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - RT associated Gender - - - - - - - - - - - - - - - Age - - 0.031 - - - - - 0.077 - 0.031 - - - 0.077 Tumor depth - - - - - - - - - - - - - - - Distant metastases 0.066 - - - 0.034 - - - - - - - 0.034 - - Stewart Treves Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - Visceral Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - 0.036 - - - - - - - - - - - - Soft tissue Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - p-value < 0.05 is considered significant ( p-value shown in bold), p-value < 0.1 is considered a trend (p-value shown);—means no significant cor - relation. Tumor depth compares deep and superficial tumors, age compares patients < 70 years of age with patients ≥ 70 years of age In UV-associated AS, some studies reveal a correla- PD-L1 expression (≥ 10(%))) according to their methyla- tion between PD-L1 positivity and worse prognosis and/ tion profile reflects with the data of Chan et al. who also or tumor cell proliferation [10, 14]. Honda et al. found an showed immunologically “cold” and “hot” clusters within association between high infiltration of PD-1 positive cells the UV-associated and other cutaneous AS of the head and favorable survival [13]. This is in accordance with and neck (n = 13) based on NanoString profiling [ 39]. The our results. One explanation could be that PD-1 expres- “hot” tumors are expected to benefit from ICI; however, sion might reflect antitumor immune response instead of to draw definitive conclusions, a thorough investigation tumor evasion. This indicates a response of the tumor cell is necessary. to high immune pressure by CD8 + T cells and could lead Recent studies have however shown that expression levels to a better prognosis as long as the balance is on the side of PD-1, PD-L1, and CD8 on their own might not be sufficient of antitumor immunity [37, 38]. to predict response to ICI [40, 41]. Other factors that may be Our finding that UV-associated AS may be classified important are tumor mutational burden (TMB), inflammation, in “cold” (low number of CD8 + T cells (< 50) and low and the further composition of the tumor immune microenvi- or no PD-1 / PD-L1 expression (< 10(%)) and “hot” AS ronment [42, 43]. It will be of (therapeutic) interest whether (high number of CD8 + T cells (≥ 50) and high PD-1/ these markers also play a role in angiosarcomas. 1 3 266 Immunologic Research (2022) 70:256–268 In the study of Chan et al., cases with high TMB (n = 3) stable) cluster. Given the scarce treatment options in AS, our were all present in the immune “hot” cluster [39]. According results provide a rationale for the future investigation/applica- to our copy number variation data in our previous methyla- tion of immune checkpoint inhibition in AS. tion study, the immune “hot” UV-associated cluster appears Supplementary Information The online version contains supplemen- to be the population harboring chromosomal instability [33]. tary material available at https://doi. or g/10. 1007/ s12026- 021- 09259-4 . Similar to tumor mutational burden, chromosomal instabil- ity might also reflect the neoantigen load of the tumor that Author contribution All authors contributed to the study’s concep- mediates T cell responses against the tumor [44]. tion and design. Material preparation, data collection, and analysis A very recent study suggests the use of tumor DNA were performed by Tess Tomassen, Melissa Hillebrandt-Roeffen, Uta Flucke, and Yvonne Versleijen-Jonkers. The first draft of the manu - methylation profiles to predict the response to anti-PD1 script was written by Tess Tomassen, and all authors commented on inhibitors in sarcomas [40]. They included only 2 AS the manuscript. All authors read and approved the final manuscript. patients (1 breast, 1 chest wall) who did not respond to the anti-PD1 treatment. Although we do not know if the Funding This research was funded by the “Stichting Bergh in het Zadel voor de kankerbestrijding” (NL). patients in our study respond to ICI, we do see a difference in methylation profiles within the UV-associated AS which Data availability Data and material are available upon request. corresponds to different immune profiles. In order to generate a more robust way to predict the Code availability Not applicable. response to ICI, it is necessary to analyze and combine mul- tiple biomarkers and validate those in a large clinical trial. Declarations It is remarkable that all previous studies chose a cut- off value of ≥ 1% or ≥ 5% for PD-L1 positive staining. Ethics approval Ethical approval for the study was obtained from Unlike the other studies, we performed statistical analy- the local certified Medical Ethics Committee of the Radboudumc, ses on PD-L1 using different cutoff values (≥ 1%, ≥ 10%, Nijmegen, The Netherlands (file number 2016–2686). and ≥ 50%). We found no statistically relevant correla- Consent to participate Not applicable. tions using a cutoff value of ≥ 1%. This may suggest that in angiosarcomas, PD-L1 as a prognostic marker is more Consent for publication Not applicable. valuable using a cutoff value of ≥ 10% or ≥ 50% compared to ≥ 1%. The optimal cutoff value for PD-L1 is still unknown Conflict of interest The authors declare no competing interests. and seems to be tumor-specific and even antibody-specific Open Access This article is licensed under a Creative Commons Attri- [45–47]. Therefore, the optimal PD-L1 cutoff value still bution 4.0 International License, which permits use, sharing, adapta- needs to be evaluated. Furthermore, in epithelial cancer, tion, distribution and reproduction in any medium or format, as long different scores are established for PD-L1, including tumor as you give appropriate credit to the original author(s) and the source, proportion score (TPS), combined positivity score (CPS), provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are and immune cell (IC) infiltrate [ 48]. In this paper, we only included in the article's Creative Commons licence, unless indicated determined TPS. The right scoring system for AS is not yet otherwise in a credit line to the material. If material is not included in established. Although it would make sense to add the PD-L1 the article's Creative Commons licence and your intended use is not positive immune cells instead of focusing only on the tumor permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a cells, we also feel that manual counting on a stained slide copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . without other markers to distinguish the type of immune cells is not enough. It would make sense to determine these scores in a future study by using multiplex immunostaining and quantifying different types of immune cells and their References PD-L1 expression in a large collection of angiosarcomas. In conclusion, with this retrospective immunohisto- 1. Cao J, Wang J, He C, Fang M. Angiosarcoma: a review of diag- nosis and current treatment. Am J Cancer Res. 2019;9:2303–13. chemical study, we present evidence of subgroup-associated 2. Florou V, Wilky BA. Current and Future Directions for Angiosar- immune profiles of AS corresponding to pathogenesis, prog - coma Therapy. Curr Treat Options Oncol. 2018;19:14. nosis, and epigenetic mechanisms. We showed that expression 3. Young RJ, Brown NJ, Reed MW, Hughes D, Woll PJ. Angiosar- of the immunological markers PD-1 and PD-L1 was clearly coma. Lancet Oncol. 2010;11:983–91. 4. Kollar A, Jones RL, Stacchiotti S, Gelderblom H, Guida M, present in several AS subgroups besides cutaneous (UV- Grignani G, Steeghs N, Safwat A, Katz D, Duffaud F, Sleijfer S, associated) AS, with varying prognostic correlations. We con- van der Graaf WT, Touati N, Litiere S, Marreaud S, Gronchi A, firmed the existence of two different clusters within the UV- Kasper B. Pazopanib in advanced vascular sarcomas: an EORTC associated subgroup, revealing one immunologically “hot” Soft Tissue and Bone Sarcoma Group (STBSG) retrospective analysis. Acta Oncol. 2017;56:88–92. (chromosomally unstable) and one “cold” (chromosomally 1 3 Immunologic Research (2022) 70:256–268 267 5. Wagner MJ, Cranmer LD, Loggers ET, Pollack SM. Proprano- 19. Kim JR, Moon YJ, Kwon KS, Bae JS, Wagle S, Kim KM, Park lol for the treatment of vascular sarcomas. J Exp Pharmacol. HS, Lee H, Moon WS, Chung MJ, Kang MJ, Jang KY. Tumor 2018;10:51–8. infiltrating PD1-positive lymphocytes and the expression of 6. Fury MG, Antonescu CR, Van Zee KJ, Brennan MF, Maki RG. PD-L1 predict poor prognosis of soft tissue sarcomas. PLoS One. A 14-year retrospective review of angiosarcoma: clinical charac- 2013;8:e82870. teristics, prognostic factors, and treatment outcomes with surgery 20. Boxberg M, Steiger K, Lenze U, Rechl H, von Eisenhart-Rothe R, and chemotherapy. Cancer J. 2005;11:241–7. Wortler K, Weichert W, Langer R, Specht K. PD-L1 and PD-1 and 7. Fayette J, Martin E, Piperno-Neumann S, Le Cesne A, Robert characterization of tumor-infiltrating lymphocytes in high grade C, Bonvalot S, Ranchere D, Pouillart P, Coindre JM, Blay JY. sarcomas of soft tissue - prognostic implications and rationale for Angiosarcomas, a heterogeneous group of sarcomas with specific immunotherapy. Oncoimmunology. 2018;7:e1389366. behavior depending on primary site: a retrospective study of 161 21. Kosemehmetoglu K, Ozogul E, Babaoglu B, Tezel GG, Gedikoglu cases. Ann Oncol. 2007;18:2030–6. G. Programmed death ligand 1 (PD-L1) expression in malignant 8. Savina M, Le Cesne A, Blay JY, Ray-Coquard I, Mir O, Toul- mesenchymal tumors. Turk Patoloji Derg. 2017;1:192–7. monde M, Cousin S, Terrier P, Ranchere-Vince D, Meeus P, 22. Que Y, Xiao W, Guan YX, Liang Y, Yan SM, Chen HY, Li Stoeckle E, Honore C, Sargos P, Sunyach MP, Le Pechoux C, QQ, Xu BS, Zhou ZW, Zhang X. PD-L1 expression is associ- Giraud A, Bellera C, Le Loarer F, Italiano A. Patterns of care and ated with FOXP3+ regulatory T-cell infiltration of soft tissue outcomes of patients with METAstatic soft tissue SARComa in a sarcoma and poor patient prognosis. J Cancer. 2017;8:2018–25. real-life setting: the METASARC observational study. BMC Med. 23. Blessin NC, Spriestersbach P, Li W, Mandelkow T, Dum D, 2017;15:78. Simon R, Hube-Magg C, Lutz F, Viehweger F, Lennartz M, 9. Boussiotis VA. Molecular and biochemical aspects of the PD-1 Fraune C, Nickelsen V, Fehrle W, Gobel C, Weidemann S, checkpoint pathway. N Engl J Med. 2016;375:1767–78. Clauditz T, Lebok P, Moller K, Steurer S, Izbicki JR, Sauter G, 10. Kawamura A, Kawamura T, Riddell M, Hikita T, Yanagi T, Ume- Minner S, Jacobsen F, Luebke AM, Buscheck F, Hoflmayer D, mura H, Nakayama M. Regulation of programmed cell death Wilczak W, Burandt E, Hinsch A. Prevalence of CD8(+) cyto- ligand 1 expression by atypical protein kinase C lambda/iota in toxic lymphocytes in human neoplasms. Cell Oncol (Dordr). cutaneous angiosarcoma. Cancer Sci. 2019;110:1780–9. 2020;43:421–30. 11. Gambichler T, Koim S, Wrobel M, Kafferlein HU, Bruning T, 24. Orth MF, Buecklein VL, Kampmann E, Subklewe M, Noessner E, Stockfleth E, Becker JC, Lang K. Expression of programmed cell Cidre-Aranaz F, Romero-Perez L, Wehweck FS, Lindner L, Issels death proteins in Kaposi sarcoma and cutaneous angiosarcoma. J R, Kirchner T, Altendorf-Hofmann A, Grunewald TGP, Knosel Immunother. 2020;43:169–74. T. A comparative view on the expression patterns of PD-L1 and 12. Fujii H, Arakawa A, Utsumi D, Sumiyoshi S, Yamamoto Y, Kitoh PD-1 in soft tissue sarcomas. Cancer Immunol Immunother. A, Ono M, Matsumura Y, Kato M, Konishi K, Shiga T, Sano S, 2020;69:1353–62. Sakaguchi S, Miyagawa-Hayashino A, Takahashi K, Uezato H, 25. Vargas AC, Maclean FM, Sioson L, Tran D, Bonar F, Mahar A, Miyachi Y, Tanioka M. CD8(+) tumor-infiltrating lymphocytes at Cheah AL, Russell P, Grimison P, Richardson L, Gill AJ. Preva- primary sites as a possible prognostic factor of cutaneous angio- lence of PD-L1 expression in matched recurrent and/or metastatic sarcoma. Int J Cancer. 2014;134:2393–402. sarcoma samples and in a range of selected sarcomas subtypes. 13. Honda Y, Otsuka A, Ono S, Yamamoto Y, Seidel JA, Morita PLoS One. 2020;15:e0222551. S, Hirata M, Kataoka TR, Takenouchi T, Fujii K, Kanekura T, 26. Lee JB, Ahn BC, Kim SH, Lee YH, Han JW, Jeon MK, Kim SH, Okubo Y, Takahashi K, Yanagi T, Hoshina D, Hata H, Abe R, Kim HS. Prognostic implications of PD-L1 expression in patients Fujimura T, Funakoshi T, Yoshino K, Masuzawa M, Amoh Y, with angiosarcoma. Future Sci OA. 2021;7(5):FSO691. Tanaka R, Fujisawa Y, Honda T, Kabashima K. Infiltration of 27. Florou V, Rosenberg AE, Wieder E, Komanduri KV, Kolonias D, PD-1-positive cells in combination with tumor site PD-L1 expres- Uduman M, Castle JC, Buell JS, Trent JC, Wilky BA. Angiosar- sion is a positive prognostic factor in cutaneous angiosarcoma. coma patients treated with immune checkpoint inhibitors: a case Oncoimmunology. 2017;6:e1253657. series of seven patients from a single institution. J Immunother 14. Shimizu A, Kaira K, Okubo Y, Utsumi D, Yasuda M, Asao T, Cancer. 2019;7:213. Nishiyama M, Takahashi K, Ishikawa O. Positive PD-L1 expres- 28. Sindhu S, Gimber LH, Cranmer L, McBride A, Kraft AS. Angio- sion predicts worse outcome in cutaneous angiosarcoma. J Glob sarcoma treated successfully with anti-PD-1 therapy - a case Oncol. 2016;3(4):360–9. report. J Immunother Cancer. 2017;5:58. 15. Bagaria SP, Chang YH, Gray RJ, Ashman JB, Attia S, Wasif N. 29. Xu F, Zheng J, Fu M, Zhou H. Antiprogrammed cell death pro- Improving long-term outcomes for patients with extra-abdom- tein 1 immunotherapy for angiosarcoma with high programmed inal soft tissue sarcoma regionalization to high-volume cent- death-ligand 1 expression: a case report. Immunotherapy. ers, improved compliance with guidelines or both? Sarcoma. 2020;12:771–6. 2018;2018:8141056. 30. Hamacher R, Kampfe D, Reuter-Jessen K, Pottgen C, Podleska 16. Botti G, Scognamiglio G, Marra L, Pizzolorusso A, Di Bonito LE, Farzaliyev F, Steinau HU, Schuler M, Schildhaus HU, Bauer M, De Cecio R, Cantile M, De Chiara A. Programmed death S. dramatic response of a PD-L1-positive advanced angiosarcoma ligand 1 (PD-L1) expression in primary angiosarcoma. J Cancer. of the scalp to pembrolizumab. Jco Precis Oncol. 2018;2:1–7. 2017;8:3166–72. 31. Momen S, Fassihi H, Davies HR, Nikolaou C, Degasperi A, 17. Googe PB, Flores K, Jenkins F, Merritt B, Moschos SJ, Grilley- Stefanato CM, Dias JML, Dasgupta D, Craythorne E, Sarkany R, Olson JE. Immune checkpoint markers in superficial angiosarco - Papa S, Nik-Zainal S. Dramatic response of metastatic cutaneous mas: PD-L1, PD-1, CD8, LAG-3, and tumor-infiltrating lympho - angiosarcoma to an immune checkpoint inhibitor in a patient with cytes. Am J Dermatopathol. 2020. xeroderma pigmentosum: whole-genome sequencing aids treat- 18. D’Angelo SP, Shoushtari AN, Agaram NP, Kuk D, Qin LX, ment decision in end-stage disease. Csh Mol Case Stud. 2019;5. Carvajal RD, Dickson MA, Gounder M, Keohan ML, Schwartz 32. Painter CA, Jain E, Tomson BN, Dunphy M, Stoddard RE, GK, Tap WD. Prevalence of tumor-infiltrating lymphocytes and Thomas BS, Damon AL, Shah S, Kim D, Gomez Tejeda Zanudo PD-L1 expression in the soft tissue sarcoma microenvironment. J, Hornick JL, Chen YL, Merriam P, Raut CP, Demetri GD, Van Hum Pathol. 2015;46(3):357–65. Tine BA, Lander ES, Golub TR, Wagle N. The Angiosarcoma 1 3 268 Immunologic Research (2022) 70:256–268 Project: enabling genomic and clinical discoveries in a rare cancer inhibitor monotherapy in sarcoma patients. J Immunother Cancer. through patient-partnered research. Nat Med. 2020;26:181–7. 2021;9. 33. Weidema ME, van de Geer E, Koelsche C, Desar IME, Kemme- 41. Petitprez F, Meylan M, de Reynies A, Sautes-Fridman C, Fridman ren P, Hillebrandt-Roeffen MHS, Ho VKY, van der Graaf WTA, WH. The Tumor microenvironment in the response to immune Versleijen-Jonkers YMH, von Deimling A, Flucke UE, PGroup. checkpoint blockade therapies. Front Immunol. 2020;11:784. DNA methylation profiling identifies distinct clusters in angiosar - 42. Meyers DE, Banerji S. Biomarkers of immune checkpoint inhibi- comas. Clin Cancer Res. 2020;26(93):100. tor efficacy in cancer. Curr Oncol. 2020;27:S106–14. 34. Casparie M, Tiebosch AT, Burger G, Blauwgeers H, van de Pol A, 43. Petitprez F, de Reynies A, Keung EZ, Chen TW, Sun CM, Calde- van Krieken JH, Meijer GA. Pathology databanking and biobank- raro J, Jeng YM, Hsiao LP, Lacroix L, Bougouin A, Moreira M, ing in The Netherlands, a central role for PALGA, the nationwide Lacroix G, Natario I, Adam J, Lucchesi C, Laizet YH, Toulmonde histopathology and cytopathology data network and archive. Cell M, Burgess MA, Bolejack V, Reinke D, Wani KM, Wang WL, Oncol. 2007;29:19–24. Lazar AJ, Roland CL, Wargo JA, Italiano A, Sautes-Fridman C, 35. Weidema ME, Flucke UE, van der Graaf WTA, Ho VKY, Hille- Tawbi HA, Fridman WH. B cells are associated with survival and brandt-Roeffen MHS, Dutch Nationwide N, Registry of H, Cyto - immunotherapy response in sarcoma. Nature. 2020;577:556–60. pathology G, Versleijen-Jonkers YMH, Husson O, Desar IME. 44. Li J, Wei Q, Wu X, Sima J, Xu Q, Wu M, Wang F, Mou H, Hu H, Prognostic Factors in a Large nationwide cohort of histologi- Zhao J, Li D, Hu J, Zhang L, Zhu X, Chen L, Luo C, Yan J, He J, cally confirmed primary and secondary angiosarcomas. Cancers Ma Y, Shao Y, Wu W, Ying J. Integrative clinical and molecular (Basel). 2019;11. analysis of advanced biliary tract cancers on immune checkpoint 36. van Erp AEM, Versleijen-Jonkers YMH, Hillebrandt-Roeffen blockade reveals potential markers of response. Clin Transl Med. MHS, van Houdt L, Gorris MAJ, van Dam LS, Mentzel T, Wei- 2020;10:e118. dema ME, Savci-Heijink CD, Desar IME, Merks HHM, van 45. Borghaei H, Paz-Ares L, Horn L, Spigel DR, Steins M, Ready Noesel MM, Shipley J, van der Graaf WTA, Flucke UE, Meyer- NE, Chow LQ, Vokes EE, Felip E, Holgado E, Barlesi F, Kohl- Wentrup FAG. Expression and clinical association of programmed haufl M, Arrieta O, Burgio MA, Fayette J, Lena H, Poddubs - cell death-1, programmed death-ligand-1 and CD8(+) lympho- kaya E, Gerber DE, Gettinger SN, Rudin CM, Rizvi N, Crino cytes in primary sarcomas is subtype dependent. Oncotarget. L, Blumenschein GR Jr, Antonia SJ, Dorange C, Harbison CT, 2017;8:71371–84. Graf Finckenstein F, Brahmer JR. Nivolumab versus Docetaxel 37. Cooper WA, Tran T, Vilain RE, Madore J, Selinger CI, Kohonen- in Advanced nonsquamous non-small-cell lung cancer. N Engl J Corish M, Yip P, Yu B, O’Toole SA, McCaughan BC, Yearley JH, Med. 2015;373:1627–39. Horvath LG, Kao S, Boyer M, Scolyer RA. PD-L1 expression is 46. Garon EB, Rizvi NA, Hui R, Leighl N, Balmanoukian AS, Eder a favorable prognostic factor in early stage non-small cell carci- JP, Patnaik A, Aggarwal C, Gubens M, Horn L, Carcereny E, noma. Lung cancer (Amsterdam, Netherlands). 2015;89:181–8. Ahn MJ, Felip E, Lee JS, Hellmann MD, Hamid O, Goldman JW, 38. Schalper KA, Velcheti V, Carvajal D, Wimberly H, Brown J, Pusz- Soria JC, Dolled-Filhart M, Rutledge RZ, Zhang J, Lunceford JK, tai L, Rimm DL. In situ tumor PD-L1 mRNA expression is associ- Rangwala R, Lubiniecki GM, Roach C, Emancipator K, Gandhi L, ated with increased TILs and better outcome in breast carcinomas. Investigators K. Pembrolizumab for the treatment of non-small- Clin Cancer Res. 2014;20:2773–82. cell lung cancer. N Engl J Med. 2015;372:2018–28. 39. Chan JY, Lim JQ, Yeong J, Ravi V, Guan P, Boot A, Tay TKY, 47. Zajac M, Ye J, Mukhopadhyay P, Jin X, Ben Y, Antal J, Gupta Selvarajan S, Md Nasir ND, Loh JH, Ong CK, Huang D, Tan J, Li AK, Rebelatto MC, Williams JA, Walker J. Optimal PD-L1-high Z, Ng CC, Tan TT, Masuzawa M, Sung KW, Farid M, Quek RHH, cutoff for association with overall survival in patients with urothe - Tan NC, Teo MCC, Rozen SG, Tan P, Futreal A, Teh BT, Soo KC. lial cancer treated with durvalumab monotherapy. PLoS One. Multiomic analysis and immunoprofiling reveal distinct subtypes 2020;15:e0231936. of human angiosarcoma. J Clin Invest. 2020;130(11):5833–46. 48. Schildhaus HU. Predictive value of PD-L1 diagnostics. Pathologe. 40. Starzer AM, Berghoff AS, Hamacher R, Tomasich E, Feldmann 2018;39:498–519. K, Hatziioannou T, Traint S, Lamm W, Noebauer-Huhmann IM, Furtner J, Mullauer L, Amann G, Bauer S, Schildhaus HU, Publisher's note Springer Nature remains neutral with regard to Preusser M, Heller G, Brodowicz T. Tumor DNA methylation jurisdictional claims in published maps and institutional affiliations. profiles correlate with response to anti-PD-1 immune checkpoint 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Immunologic Research Springer Journals

Analysis of PD-1, PD-L1, and T-cell infiltration in angiosarcoma pathogenetic subgroups

Loading next page...
 
/lp/springer-journals/analysis-of-pd-1-pd-l1-and-t-cell-infiltration-in-angiosarcoma-SNNhZtz4E4

References (64)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
ISSN
0257-277X
eISSN
1559-0755
DOI
10.1007/s12026-021-09259-4
Publisher site
See Article on Publisher Site

Abstract

Angiosarcoma (AS) is a rare malignancy with a poor prognosis. It can develop spontaneously or due to previous radiotherapy (RT), ultraviolet (UV) radiation, or lymphoedema (Stewart Treves AS). Novel therapeutic approaches are needed, but progress is hindered because of the heterogeneity and rarity of AS. In order to explore the potential of immune checkpoint inhibition (ICI), we investigated the protein expression of programmed cell death 1 (PD-1), programmed death-ligand 1 (PD-L1), and CD8 + T cells in 165 AS cases in relation to AS subgroups based on clinical classification and in relation to whole-genome methylation profiling based clusters (A1, A2, B1, B2). High PD-L1 and PD-1 expression were predominantly shown in UV-associated, visceral, and soft tissue AS. RT-associated AS showed predominantly high PD-1 expression. CD8 + T cell infiltration w as present in the majority of AS samples. Within the UV-associated AS, two different clusters can be distinguished by DNA methylation prol fi ing. Cases in cluster A1 showed higher PD-1 (p = 0.015), PD-L1 (p = 0.015), and CD8 + T cells (p = 0.008) compared to those in cluster B2, suggesting that these UV-AS tumors are more immunogenic than B2 tumors showing a difference even within one subgroup. In soft tissue AS, combined PD-1 and PD-L1 expression showed a trend toward poor survival (p = 0.051), whereas in UV-associated AS, PD-1 expression correlated with better survival (p = 0.035). In conclusion, we show the presence of PD-1, PD-L1, and CD8 + T cells in the majority of AS but reveal differ - ences between and within AS subgroups, providing prognostic information and indicating to be predictive for ICI. Keywords Angiosarcoma · Subgroups · Programmed cell death 1 · Programmed death-ligand 1 (PD-L1) · Immune checkpoint inhibition Introduction which DNA damaging factors including radiation, UV light exposure or chronic lymphoedema play an important role AS is a rare and aggressive vasoformative sarcoma aris- [1–3]. Current treatment options include surgery, RT, and/ ing at different anatomical sites, including skin, soft tissue, or chemotherapy, depending on the extent of the disease. In bone, and visceral organs. AS can be clinically classified into addition, the multi-tyrosine kinase inhibitor pazopanib is also primary AS (with unknown etiology) or secondary AS, in applied in daily practice [4] and treatment with the generic ß-blocker propranolol has been suggested [5]. The survival of AS patients is poor with a reported 5-year survival of *Among others; Jos W.R. Meyer, Rijnstate Hospital, Arnhem, The only 30–40% [6–8], emphasizing the need for novel treat- Netherlands and Marieke C.H. Hogenes, LABPON, Hengelo, The Netherlands. ment options. A potential approach for AS treatment is ICI. Tumor cells * Yvonne M. H. Versleijen-Jonkers can upregulate PD-L1 on their membrane to promote immune yvonne.versleijen-jonkers@radboudumc.nl suppression. Interaction with the receptor programmed cell Department of Pathology, Radboud University Medical death 1 (PD-1) on CD8 + T cells renders the T cell inactive, Center, Nijmegen, The Netherlands and thus prevents the killing of tumor cells. ICI with anti- Department of Medical Oncology (Internal Postal Code: PD-1 antibodies can reactivate the cytotoxic function of T 452), Radboud University Medical Center, P.O. Box 9101, cells leading to the subsequent killing of tumor cells [9]. 6500 HB Nijmegen, The Netherlands In order to determine the role of ICI in AS, it is necessary Princess Máxima Center for Pediatric Oncology, Utrecht, to examine the expression of PD-L1, PD-1, and the presence The Netherlands 1 3 Immunologic Research (2022) 70:256–268 257 of CD8 + T cells as potential biomarkers in this respect. Sev- pathogenetic subgroups based on available clinical data and eral studies investigated the expression of these biomarkers pathology reports [35]. in AS, reporting variable levels of expression and varying Tumor samples were collected on tissue microarrays correlations with prognosis (shown in Table  1). Of note, (TMAs) and divided into the different subgroups, includ - most studies were performed on cutaneous (predominantly ing 44 UV-associated, 14 cutaneous not UV-associated, 55 UV-associated) AS or small numbers of other subtypes, RT-associated, 14 Stewart Treves (lymphoedema-associated often not further specified. cases), 27 visceral, and 11 soft tissue cases. Only primary Clinical data with regard to the application of ICI in AS is localized tumor samples were selected for this study. Of limited to small case series. In one of these case series, three these samples, 33 have been previously subject to DNA UV-associated AS were treated with anti-PD-1 (one in com- methylation profiling [ 33]. bination with CTLA-4 inhibition) showing partial response. One RT-associated AS also showed partial response (with Clinical data anti-PD1) and one primary breast AS showed progressive disease on axitinib (VEGFR inhibitor) combined with PD-1 Clinical data were received from the nationwide Netherlands inhibition [27]. Four patients suffering from UV-associated Cancer Registry and were linked to data from the Dutch AS with significant PD-L1 expression were successfully pathology registry (PALGA). Ethical approval for the study treated with anti-PD-1 [28–31]. In the recent AS patient- was obtained from the local certified Medical Ethics Com - partnered-genomic study of Painter et  al., a total of six mittee of the Radboudumc, Nijmegen, The Netherlands (file patients were treated with anti-PD-1 [32]. Of these patients, number 2016–2686). two out of three UV-associated AS with high mutational burden (> 150 mutations/Mb) showed a complete response, Immunohistochemistry whereas no clinical benefit was found in the other three patients (all non-UV AS with low mutational burden) [32]. Immunohistochemical analysis was performed to inves- Pecora et al. published a case of primary AS of the temple tigate PD-1 and PD-L1 expression and the presence of and one RT-associated AS with complete clinical remission CD8 + lymphocytes in the tumor. Tonsil (PD-L1 + , PD-1 + , on combined anti-CTLA4 and anti-PD-1 therapy (Pecora and CD8 +) and appendix (CD8 + and PD1 +) served as et al. CTOS2019). positive controls. Immunohistochemistry was performed Taken together, the previously mentioned studies sug- on 4-µm-thick FFPE sections of AS TMAs with one or two gest a potential therapeutic role for ICI in especially UV- cores per sample from representative tumor areas (core size associated AS. However, it yet remains unclear whether 2 mm) to allow simultaneous examination of patient speci- immunotherapy could be of interest for all AS patients or mens under identical conditions. Staining was performed in only for certain pathogenetic subgroups. Our recent DNA the Lab Vision Autostainer 360 (Thermo Fisher Scientific) methylation profiling study confirmed the existence of these by using the EnVision FLEX, pH high Link kit (Dako), and subgroups of AS on an epigenetic level, which did not fully monoclonal rabbit anti-PD-L1 (1:800, clone E1L3N, Cell match the clinical subtypes [33]. Signaling Technology), monoclonal mouse anti-PD-1 (1:20, In the current retrospective study, we aimed to character- clone MRQ-22, Cell Marque) or monoclonal mouse anti- ize PD-1, PD-L1 expression, and the presence of CD8 + T CD8 (1:80, clone C8/144B, Dako). cells in a large cohort of AS tumor samples and their prog- PD-L1 expression on the tumor cells was scored as 0% nostic relevance to further explore the heterogeneity and the ( −), 1–10% (+ / −), 10–50% ( +) or ≥ 50% positive tumor need to differentiate between the different AS subgroups. cells (+ +). All CD8 and PD-1 positive T cells were counted and subdivided in three categories: < 10 ( −), 10–50 ( +), or ≥ 50 positive cells (+ +) per tumor core [36]. Digital images were generated with VisionTekTM Materials and methods (Sakura, version 2.6) and analyzed at × 20 magnification. Tumor sample collection Statistical analysis We collected formalin-fixed paraffin-embedded (FFPE) Statistical analyses were performed using IBM SPSS Sta- tumor tissue of AS patients by a nationwide search through tistics 25. p-values < 0.05 were considered significant and PALGA (Dutch nationwide network and registry of histo- p-values < 0.1 were considered a trend. Relations between and cytopathology) diagnosed between 1989 and 2015 in categorical parameters were assessed by chi-square or Fish- the Netherlands [34]. All cases were reviewed by an expert er’s exact testing as appropriate, and associations with over- pathologist (UF), and confirmed AS cases were divided into all survival (OS) were assessed by the Kaplan–Meier method 1 3 258 Immunologic Research (2022) 70:256–268 1 3 Table 1 Overview of studies regarding the presence of PD-1, PD-L1, and CD8 in AS Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Kawamura 29 Cutaneous 12 (41%) 22 (76%) 10 (34%) > 5% Mouse anti- Rabbit anti- PD-L1 et al. 2019 human ab human ab expression [10] (CST) (CST) associated with poor survival Gambichler 12 Cutaneous 4 (33%) 5 (42%) Ab92484 Ab205921 et al. 2020 (Abcam) (Abcam) [11] Fujii et al. 55 Cutaneous 20/40 (50%) ≥ Median/ C8/144B High number 2014 [12] (mainly HPF (Dako) of CD8 in UV) primary tumor cor- related with improved survival Honda et al. 106 Cutaneous 19 (18%) 32 (30%) 9 (9%) > 50/3HPF > 5% NAT105 SP142 PD-1 + cells 2017 [13] (mainly (Abcam) (Spring (especially UV) Biosience) in combina- tion with PD-L1 expression) corre- lated with improved survival Shimizu et 52 Cutaneous 21 (40%) 24 (46%) ≥ 5% > Median Rabbit mon- Abcam PD-L1 and al. 2017 (mainly oclonal CD8 are [14] UV) antibody associ- (Abcam) ated with a worse outcome 3 3 2 Bagaria et 26 Cutaneous 1 (4%) 5 (19%) ≥ 5% NAT105 SP-142 and No al. 2018 (19), other (Ventana) SP-263 [15] (8) (Ventana) Immunologic Research (2022) 70:256–268 259 1 3 Table 1 (continued) Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Botti et al. 24 Primary AS 16 (66%) ≥ 5% SP-142 No 2017 [16] (7 breast, (Spring 5 soft Biosience) tissue, 4 bone, 4 skin, 4 visceral) Googe et al. 10 Skin (9), 10 (100%) 10 (100%) 10 (100%) 10 (100%) Low: ≥ 1% NAT105 ZR 3 4B11 (Leica 2020 [17] soft tissue High: ≥ 50% (Cell (Cell Biosys- (1) Marque) Marque) tems) D’Angelo et 3 n.s 0 (0%) 1 (33%) ≥ 1% ≥ 5% Dako C8/144B al. 2015 (Dako) [18] Kim et al. 5 n.s 4 (80%) 4 (80%) 4 (80%) ≥ 1 Total NAT H-130 2013 [19] score ≥ 8 (Abcam) (SCB) Boxberg et 23 n.s 8 (35%) ≥ 1% SP263 (Ven- al. 2018 4 (17%) ≥ 5% tana) [20] 3 (13%) ≥ 10% 2 (9%) ≥ 50% Kosemehme- 7 n.s 1 (14%) > 5% E1L3N toglu et al. (CST) 2017 [21] Que et al. 5 n.s 1 (20%) > 1% E1L3N 2017 [22] (CST) Blessin et al. 25 n.s Median TC8 2020 [23] count 95 (ONCOdi- cells/mm anova) Orth et al. 6 n.s 2/5 3 (50%) 1 (17%) ≥ 4/HPF > 1% 315 M (Cell E1L3N 2020 [24] (40%) Marque) (CST) Vargas et al. 17 n.s 5 (29%) ≥ 1% SP263 (Ven- 2020 [25] tana) Lee et al. 70 n.s 13 (19%) Combined 22C3 PD-L1 2021 [26] positive (Agilent expression score ≥ 1 Technolo- correlated gies) with poor survival in metastatic AS patients 260 Immunologic Research (2022) 70:256–268 1 3 Table 1 (continued) Study refer- N AS subtype Protein expression Threshold for positivity Antibodies Correlation ence with prog- PD-1 (%) PD-L1 (%) PD-1 CD8 (%) PD-1 PD-L1 CD8 PD-1 PD-L1 CD8 nosis andPD-L1 (%) Summary of N AS subtype PD-1 PD-L1 PD-1 and CD8 PD-1 PD-L1 CD8 results expres- expres- PD-L1 expres- threshold threshold threshold sion (%) sion (%) expres- sion (%) sion (%) 286 Cutaneous 44/156 92/212 28/144 53/101 > 50/3HPF ≥ 1% ≥ Median (28%) (43%) (19%) (52%) to ≥ 5% 7 Breast 4/7 (57%) ≥ 5% 6 Soft tissue 1/1 (100%) 5/6 (83%) 1/1 (100%) 1/1 (100%) ≥ 1% to ≥ 5% 4 Bone 4/4 (100%) ≥ 5% 4 Visceral 1/4 (25%) ≥ 5% 169 n.s 6/10 (60%) 35/136 5/11 (45%) 1/3 (33%) ≥ 1 to ≥ 4/ ≥ 1% to ≥ 8 ≥ 5% (26%) HPF Abbreviations: AS, angiosarcoma; CST, Cell Signaling Technology, HPF, high power field; n.s., not specified; SCB, Santa Cruz Biotechnology; UV, UV-associated AS; For PD-L1, the staining intensity score was classified as 0 (no staining), 1 (weak staining), and 2 (intermediate staining), and 3 (strong staining). The area of staining was scored as 0 (0–10% of the cells stained), 1 (11–33% of the cells stained), 2 (34–66% of the cells stained), and 3 (67–100% of the cells stained). The total score was determined as the sum of the intensity score and 2 3 4 the staining proportion score of two different TMAs. The total score ≥ 8 was determined as positive. Membranous expression of tumor cells. Expression not displayed per subtype. Results are summarized per subtype Immunologic Research (2022) 70:256–268 261 with the logrank test. Tumors positive for one marker or between the different levels of expression was made in the a combination of markers were compared to tumors nega- analysis. tive for that particular marker or combination. A distinction Results Table 2 Patient characteristics Immune profiles in AS subtypes N (%) AS samples 165 The expression of PD-1, PD-L1, and CD8 was assessed in UV associated 44 (27) 165 AS samples divided over 6 different subgroups. Patient Cutaneous not UV associated 14 (8) characteristics are shown in Table  2. Staining results are RT associated 55 (33) presented in Table 3 with an example of each staining shown Stewart Treves 14 (8) in Supplementary Fig. 1. Visceral 27 (16) High PD-1 and PD-L1 expressions were predominantly Soft tissue 11 (7) seen in clinically defined soft tissue (40%), UV-associated Extent of disease (18%), and visceral (17%) AS subgroups. Besides, RT-   Localized 80 (48) associated AS showed predominantly high PD-1 expression   Invasion adjacent structures 9 (5) (32%). Infiltration of high numbers of CD8 + T cells was   Lymph node involvement 4 (2) present in the majority of AS samples across all different   Distant metastases 14 (8) subgroups (64–86%) (Table 3).   Unknown 58 (35) The two main clusters (A and B) defined by our previ - Tumor depth ous genome-wide array-based DNA methylation profiling   Superficial 44 (27) study were each subdivided into 2 separate clusters (A1,   Deep 5 (3) A2 and B1, B2) (Fig. 1A). Cluster A1 consisted exclusively   Unknown 116 (70) of UV-associated cases, whereas A2 primarily consisted of Distant metastases RT-associated cases. Cluster B1 had both visceral and soft   No 111 (67) tissue cases, and cluster B2 was mixed, including cases of   Yes 20 (12) UV-associated AS.   Unknown 34 (21) In the current study, a significantly higher PD-L1 expres - Age sion (≥ 10%) and PD-1 expression (≥ 10) was found in the    < 40 6 (4) UV-associated cases in cluster A1 versus those in cluster B2    ≥ 40 < 70 55 (33) (for both stainings 6/7 (86%) in A1 versus 0/4 (0%) in B2,    ≥ 70 104 (63) p = 0.015) (Fig. 1B). Gender A high amount of CD8-positive T cells (≥ 50) was   Male 54 (33) observed in all UV-associated cases in cluster A1 versus   Female 111 (67) none of the UV-associated cases in cluster B2 (7/7 (100%) Follow-up status in A1 versus 0/3 (0%) in B2 (1 case was not evaluable), (median follow-up 14.8 months) p = 0.008) (Fig. 1B). A difference in the number of copy   Alive 20 (12) number variations (CNVs) was seen between the UV-associ-   Deceased 145 (88) ated AS cases in cluster A1 versus B2, as already described Table 3 PD-L1, PD-1 expression, and the presence of CD8 + T cells in angiosarcoma AS subgroup N PD-L1 ≥ 1% PD-L1 ≥ 10% PD-L1 ≥ 50% PD-1 ≥ 10 PD1 ≥ 50 CD8 ≥ 10 CD8 ≥ 50 PD-1 and PD-1 and PD-L1 ≥ 10 PD-L1 ≥ 50 (%) (%) UV associated 44 89% 66% 37% 66% 39% 98% 79% 50% 18% Cutaneous not UV 14 86% 50% 7% 64% 7% 93% 64% 21% 0% RT associated 55 79% 52% 8% 68% 32% 98% 81% 39% 6% Stewart Treves 14 75% 50% 8% 43% 7% 100% 86% 36% 0% Visceral 27 79% 54% 46% 52% 20% 92% 69% 38% 17% Soft tissue 11 100% 78% 56% 72% 36% 91% 82% 60% 40% 1 3 262 Immunologic Research (2022) 70:256–268 Fig. 1 Overview of the division of AS subgroups over the dif- ferent methylation clusters (A). Expression of PD-1, PD-L1, CD8, and the presence of copy number variations (CNVs) in UV-associated AS in cluster A1 versus B2 (B) A1 A2 B1 B2 Cluster A1 Cluster B2 PD-1 14% <10 43% ≥10<50 100% ≥50 43% PD-L1 14% <10 43% 50% 50% ≥10<50 ≥50 43% CD8 <10 25% ≥10<50 ≥50 100% unknown 75% CNVs <10 25% ≥10<20 ≥20 75% 100% 1 3 Immunologic Research (2022) 70:256–268 263 in our methylation profiling study (mean number of CNVs (19%) female, p = 0.039). Also, combined positive PD-L1 34.4 (range 23–39) in cluster A1 versus 7.3 (range 3–17) in and CD8 (≥ 10(%)) expression and high PD-L1 and CD8 cluster B2, p < 0.001) (Fig. 1B). (≥ 50(%)) expression showed a trend toward a correlation For the other AS subgroups, no significant differences in with the male gender (p = 0.070/0.075). High PD-L1 expres- expression of PD-L1, PD-1, or CD8 were observed between sion (≥ 50%) was also more common in deep tumors (3/4 clusters (Supplementary Fig. 2). (75%) deep versus 6/40 (15%) superficial tumors, p = 0.023). In RT-associated AS, PD-L1 expression (≥ 1%) showed Prognostic relevance of the immune profile a trend toward a correlation with the absence of metastases (p = 0.066), whereas no significant correlation with the pres - In Supplementary Table 1, we present the univariate analysis ence of distant metastases was observed at the thresholds of associations of the expression of the different markers of 10 or 50%. High PD-L1 expression (≥ 50%) did show with overall survival. In the total group PD-1, PD-L1, CD8, a significant correlation with the age below 70 years (4/21 or combined expression did not significantly correlate with (19%) < 70  years of age versus 0/27 (0%) ≥ 70  years of survival, although we did observe a trend toward a worse age, p = 0.031). All cases with high PD-L1 expression also overall survival for patients with ≥ 10% PD-L1 expres- showed a high CD8 expression. The combination of both sion in their tumor versus < 10% PD-L1 (median 11.0 ± 2.4 high PD-L1 and PD-1 expressions, as well as the combina- versus 17.1 ± 4.1  months, respectively, p = 0.088) and for tion of high PD-L1, PD-1, and CD8 expressions showed a patients with both ≥ 10% PD-L1 expression and ≥ 10 CD8- trend toward a correlation with the age group below 70 years positive T cells in their tumor versus those with no expres- (p = 0.077). High PD-1 expression (≥ 50) correlated with sion of both markers (median 11.0 ± 2.4  months versus the presence of distant metastases (PD-1 ≥ 50 in 3/3 (100%) 18.4 ± 4.1 months, respectively, p = 0.083) (Fig. 2A). patients with distant metastases versus 11/27 (41%) without In the cutaneous not UV-associated, RT-associated, Stew- distant metastases, p = 0.034). Cases with high PD-1 expres- art Treves, and visceral AS groups, no significant correla - sion also showed high CD8 expression. tions with survival were observed. We excluded correlations In visceral AS, high PD-L1 expression (≥ 50%) corre- when groups of only one patient were involved. lated with the absence of distant metastases (PD-L1 ≥ 50% in In patients with soft tissue AS, the presence of both PD-1 0/5 (0%) patients with distant metastases versus 8/13 (62%) (≥ 10) and PD-L1 (≥ 10%) showed a trend toward poor sur- without distant metastases, p = 0.036). vival (estimate mean survival time 6.8 ± 2.7 months (both positive, n = 6) versus 145.5 ± 41.3 months (not both posi- tive, n = 4), p = 0.051) (Fig. 2B). Discussion In UV-associated AS, the presence of PD-1 positive cells in the tumor (≥ 10) correlated with better overall This is the first study mapping the immunological landscape survival (median 20.6 ± 8.5  months for PD1 ≥ 10 versus in different AS subgroups as well as in genome-wide meth - 5.8 ± 2.1 months for PD1 < 10, p = 0.035). The same applies ylation profiling clusters. We detected relevant differences to the presence of high numbers of PD-1 positive cells in between the various subgroups. the tumor (≥ 50) (median 22.1 ± 16.7 months for PD1 ≥ 50 We showed a high expression of both PD-1 and PD-L1 versus 8.3 ± 2.8 months for PD1 < 50, p = 0.020) (Fig. 2C). predominantly in UV-associated, visceral, and soft tissue All PD-1 positive tumors in this subgroup were also CD8 AS subgroups and high PD-1 expression in RT-associated positive. AS, whereas infiltration of CD8 + T cells was present in the UV-associated AS patients in cluster A1 (instable, more majority of AS samples. In soft tissue AS the presence of immunogenic (“hot”) cluster, n = 7) showed a trend toward both PD-1 and PD-L1 expression showed a trend toward better overall survival compared to those in cluster B2 poor survival, whereas in UV-associated AS, PD-1 expres- (stable, “cold” cluster, n = 4) (median 22.2 ± 11.4 versus sion was correlated with better survival. 1.6 ± 4.5 months, p = 0.053) (Fig. 2D). These results reflect the heterogeneity in immunological response associated with prognosis for the diverse AS sub- Correlations with patient characteristics groups and underline the need to differentiate between them. So far, most studies have analyzed PD-1, PD-L1, and CD8 Correlations with gender, age, tumor depth, and presence expression only in cutaneous, mainly UV-associated AS of distant metastases are presented in Table 4. In the total (displayed in Table 1). Based on the observed high expres- group, PD-L1 expression (≥ 10%) already showed a trend sion of PD-1, PD-L1, and CD8 in visceral and soft tissue AS toward a correlation with the male gender (p = 0.075), in the current study, ICI might also be successful in these whereas high PD-L1 expression (≥ 50%) correlated signifi - subgroups. Individual immune profiling before the start of cantly with the male gender (17/47 (36%) male versus 19/98 ICI could be considered to select more vulnerable tumors. 1 3 264 Immunologic Research (2022) 70:256–268 Fig. 2 Kaplan–Meier curves showing the significant differences and ence of CD8 + T cells in the tumor and a Kaplan–Meier curve show- trends in overall survival in A AS total group, B soft tissue AS, and C ing the difference in overall survival between UV-associated AS in UV-associated AS according to PD-1, PD-L1 expression, or the pres- cluster A1 versus B2 (D) 1 3 Immunologic Research (2022) 70:256–268 265 Table 4 Correlations between (combinations of) biomarkers and clinical data AS subgroup Clinical marker PD-L1 PD-1 CD8 PD-L1 and PD-L1 and PD-1 and PD-L1 PD-1 CD8 CD8 and PD-1 and CD8 1 10 50 10 50 10 50 10 50 10 50 10 50 10 50 All Gender - 0.075 0.039 - - - - - - 0.070 0.075 - - - - Age - - - - - - - - - - - - - - - Tumor depth - - 0.023 - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - UV associated Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - Cutaneous not UV Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - RT associated Gender - - - - - - - - - - - - - - - Age - - 0.031 - - - - - 0.077 - 0.031 - - - 0.077 Tumor depth - - - - - - - - - - - - - - - Distant metastases 0.066 - - - 0.034 - - - - - - - 0.034 - - Stewart Treves Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - Visceral Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - 0.036 - - - - - - - - - - - - Soft tissue Gender - - - - - - - - - - - - - - - Age - - - - - - - - - - - - - - - Tumor depth - - - - - - - - - - - - - - - Distant metastases - - - - - - - - - - - - - - - p-value < 0.05 is considered significant ( p-value shown in bold), p-value < 0.1 is considered a trend (p-value shown);—means no significant cor - relation. Tumor depth compares deep and superficial tumors, age compares patients < 70 years of age with patients ≥ 70 years of age In UV-associated AS, some studies reveal a correla- PD-L1 expression (≥ 10(%))) according to their methyla- tion between PD-L1 positivity and worse prognosis and/ tion profile reflects with the data of Chan et al. who also or tumor cell proliferation [10, 14]. Honda et al. found an showed immunologically “cold” and “hot” clusters within association between high infiltration of PD-1 positive cells the UV-associated and other cutaneous AS of the head and favorable survival [13]. This is in accordance with and neck (n = 13) based on NanoString profiling [ 39]. The our results. One explanation could be that PD-1 expres- “hot” tumors are expected to benefit from ICI; however, sion might reflect antitumor immune response instead of to draw definitive conclusions, a thorough investigation tumor evasion. This indicates a response of the tumor cell is necessary. to high immune pressure by CD8 + T cells and could lead Recent studies have however shown that expression levels to a better prognosis as long as the balance is on the side of PD-1, PD-L1, and CD8 on their own might not be sufficient of antitumor immunity [37, 38]. to predict response to ICI [40, 41]. Other factors that may be Our finding that UV-associated AS may be classified important are tumor mutational burden (TMB), inflammation, in “cold” (low number of CD8 + T cells (< 50) and low and the further composition of the tumor immune microenvi- or no PD-1 / PD-L1 expression (< 10(%)) and “hot” AS ronment [42, 43]. It will be of (therapeutic) interest whether (high number of CD8 + T cells (≥ 50) and high PD-1/ these markers also play a role in angiosarcomas. 1 3 266 Immunologic Research (2022) 70:256–268 In the study of Chan et al., cases with high TMB (n = 3) stable) cluster. Given the scarce treatment options in AS, our were all present in the immune “hot” cluster [39]. According results provide a rationale for the future investigation/applica- to our copy number variation data in our previous methyla- tion of immune checkpoint inhibition in AS. tion study, the immune “hot” UV-associated cluster appears Supplementary Information The online version contains supplemen- to be the population harboring chromosomal instability [33]. tary material available at https://doi. or g/10. 1007/ s12026- 021- 09259-4 . Similar to tumor mutational burden, chromosomal instabil- ity might also reflect the neoantigen load of the tumor that Author contribution All authors contributed to the study’s concep- mediates T cell responses against the tumor [44]. tion and design. Material preparation, data collection, and analysis A very recent study suggests the use of tumor DNA were performed by Tess Tomassen, Melissa Hillebrandt-Roeffen, Uta Flucke, and Yvonne Versleijen-Jonkers. The first draft of the manu - methylation profiles to predict the response to anti-PD1 script was written by Tess Tomassen, and all authors commented on inhibitors in sarcomas [40]. They included only 2 AS the manuscript. All authors read and approved the final manuscript. patients (1 breast, 1 chest wall) who did not respond to the anti-PD1 treatment. Although we do not know if the Funding This research was funded by the “Stichting Bergh in het Zadel voor de kankerbestrijding” (NL). patients in our study respond to ICI, we do see a difference in methylation profiles within the UV-associated AS which Data availability Data and material are available upon request. corresponds to different immune profiles. In order to generate a more robust way to predict the Code availability Not applicable. response to ICI, it is necessary to analyze and combine mul- tiple biomarkers and validate those in a large clinical trial. Declarations It is remarkable that all previous studies chose a cut- off value of ≥ 1% or ≥ 5% for PD-L1 positive staining. Ethics approval Ethical approval for the study was obtained from Unlike the other studies, we performed statistical analy- the local certified Medical Ethics Committee of the Radboudumc, ses on PD-L1 using different cutoff values (≥ 1%, ≥ 10%, Nijmegen, The Netherlands (file number 2016–2686). and ≥ 50%). We found no statistically relevant correla- Consent to participate Not applicable. tions using a cutoff value of ≥ 1%. This may suggest that in angiosarcomas, PD-L1 as a prognostic marker is more Consent for publication Not applicable. valuable using a cutoff value of ≥ 10% or ≥ 50% compared to ≥ 1%. The optimal cutoff value for PD-L1 is still unknown Conflict of interest The authors declare no competing interests. and seems to be tumor-specific and even antibody-specific Open Access This article is licensed under a Creative Commons Attri- [45–47]. Therefore, the optimal PD-L1 cutoff value still bution 4.0 International License, which permits use, sharing, adapta- needs to be evaluated. Furthermore, in epithelial cancer, tion, distribution and reproduction in any medium or format, as long different scores are established for PD-L1, including tumor as you give appropriate credit to the original author(s) and the source, proportion score (TPS), combined positivity score (CPS), provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are and immune cell (IC) infiltrate [ 48]. In this paper, we only included in the article's Creative Commons licence, unless indicated determined TPS. The right scoring system for AS is not yet otherwise in a credit line to the material. If material is not included in established. Although it would make sense to add the PD-L1 the article's Creative Commons licence and your intended use is not positive immune cells instead of focusing only on the tumor permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a cells, we also feel that manual counting on a stained slide copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . without other markers to distinguish the type of immune cells is not enough. It would make sense to determine these scores in a future study by using multiplex immunostaining and quantifying different types of immune cells and their References PD-L1 expression in a large collection of angiosarcomas. In conclusion, with this retrospective immunohisto- 1. Cao J, Wang J, He C, Fang M. Angiosarcoma: a review of diag- nosis and current treatment. Am J Cancer Res. 2019;9:2303–13. chemical study, we present evidence of subgroup-associated 2. Florou V, Wilky BA. Current and Future Directions for Angiosar- immune profiles of AS corresponding to pathogenesis, prog - coma Therapy. Curr Treat Options Oncol. 2018;19:14. nosis, and epigenetic mechanisms. We showed that expression 3. Young RJ, Brown NJ, Reed MW, Hughes D, Woll PJ. Angiosar- of the immunological markers PD-1 and PD-L1 was clearly coma. Lancet Oncol. 2010;11:983–91. 4. Kollar A, Jones RL, Stacchiotti S, Gelderblom H, Guida M, present in several AS subgroups besides cutaneous (UV- Grignani G, Steeghs N, Safwat A, Katz D, Duffaud F, Sleijfer S, associated) AS, with varying prognostic correlations. We con- van der Graaf WT, Touati N, Litiere S, Marreaud S, Gronchi A, firmed the existence of two different clusters within the UV- Kasper B. Pazopanib in advanced vascular sarcomas: an EORTC associated subgroup, revealing one immunologically “hot” Soft Tissue and Bone Sarcoma Group (STBSG) retrospective analysis. Acta Oncol. 2017;56:88–92. (chromosomally unstable) and one “cold” (chromosomally 1 3 Immunologic Research (2022) 70:256–268 267 5. Wagner MJ, Cranmer LD, Loggers ET, Pollack SM. Proprano- 19. Kim JR, Moon YJ, Kwon KS, Bae JS, Wagle S, Kim KM, Park lol for the treatment of vascular sarcomas. J Exp Pharmacol. HS, Lee H, Moon WS, Chung MJ, Kang MJ, Jang KY. Tumor 2018;10:51–8. infiltrating PD1-positive lymphocytes and the expression of 6. Fury MG, Antonescu CR, Van Zee KJ, Brennan MF, Maki RG. PD-L1 predict poor prognosis of soft tissue sarcomas. PLoS One. A 14-year retrospective review of angiosarcoma: clinical charac- 2013;8:e82870. teristics, prognostic factors, and treatment outcomes with surgery 20. Boxberg M, Steiger K, Lenze U, Rechl H, von Eisenhart-Rothe R, and chemotherapy. Cancer J. 2005;11:241–7. Wortler K, Weichert W, Langer R, Specht K. PD-L1 and PD-1 and 7. Fayette J, Martin E, Piperno-Neumann S, Le Cesne A, Robert characterization of tumor-infiltrating lymphocytes in high grade C, Bonvalot S, Ranchere D, Pouillart P, Coindre JM, Blay JY. sarcomas of soft tissue - prognostic implications and rationale for Angiosarcomas, a heterogeneous group of sarcomas with specific immunotherapy. Oncoimmunology. 2018;7:e1389366. behavior depending on primary site: a retrospective study of 161 21. Kosemehmetoglu K, Ozogul E, Babaoglu B, Tezel GG, Gedikoglu cases. Ann Oncol. 2007;18:2030–6. G. Programmed death ligand 1 (PD-L1) expression in malignant 8. Savina M, Le Cesne A, Blay JY, Ray-Coquard I, Mir O, Toul- mesenchymal tumors. Turk Patoloji Derg. 2017;1:192–7. monde M, Cousin S, Terrier P, Ranchere-Vince D, Meeus P, 22. Que Y, Xiao W, Guan YX, Liang Y, Yan SM, Chen HY, Li Stoeckle E, Honore C, Sargos P, Sunyach MP, Le Pechoux C, QQ, Xu BS, Zhou ZW, Zhang X. PD-L1 expression is associ- Giraud A, Bellera C, Le Loarer F, Italiano A. Patterns of care and ated with FOXP3+ regulatory T-cell infiltration of soft tissue outcomes of patients with METAstatic soft tissue SARComa in a sarcoma and poor patient prognosis. J Cancer. 2017;8:2018–25. real-life setting: the METASARC observational study. BMC Med. 23. Blessin NC, Spriestersbach P, Li W, Mandelkow T, Dum D, 2017;15:78. Simon R, Hube-Magg C, Lutz F, Viehweger F, Lennartz M, 9. Boussiotis VA. Molecular and biochemical aspects of the PD-1 Fraune C, Nickelsen V, Fehrle W, Gobel C, Weidemann S, checkpoint pathway. N Engl J Med. 2016;375:1767–78. Clauditz T, Lebok P, Moller K, Steurer S, Izbicki JR, Sauter G, 10. Kawamura A, Kawamura T, Riddell M, Hikita T, Yanagi T, Ume- Minner S, Jacobsen F, Luebke AM, Buscheck F, Hoflmayer D, mura H, Nakayama M. Regulation of programmed cell death Wilczak W, Burandt E, Hinsch A. Prevalence of CD8(+) cyto- ligand 1 expression by atypical protein kinase C lambda/iota in toxic lymphocytes in human neoplasms. Cell Oncol (Dordr). cutaneous angiosarcoma. Cancer Sci. 2019;110:1780–9. 2020;43:421–30. 11. Gambichler T, Koim S, Wrobel M, Kafferlein HU, Bruning T, 24. Orth MF, Buecklein VL, Kampmann E, Subklewe M, Noessner E, Stockfleth E, Becker JC, Lang K. Expression of programmed cell Cidre-Aranaz F, Romero-Perez L, Wehweck FS, Lindner L, Issels death proteins in Kaposi sarcoma and cutaneous angiosarcoma. J R, Kirchner T, Altendorf-Hofmann A, Grunewald TGP, Knosel Immunother. 2020;43:169–74. T. A comparative view on the expression patterns of PD-L1 and 12. Fujii H, Arakawa A, Utsumi D, Sumiyoshi S, Yamamoto Y, Kitoh PD-1 in soft tissue sarcomas. Cancer Immunol Immunother. A, Ono M, Matsumura Y, Kato M, Konishi K, Shiga T, Sano S, 2020;69:1353–62. Sakaguchi S, Miyagawa-Hayashino A, Takahashi K, Uezato H, 25. Vargas AC, Maclean FM, Sioson L, Tran D, Bonar F, Mahar A, Miyachi Y, Tanioka M. CD8(+) tumor-infiltrating lymphocytes at Cheah AL, Russell P, Grimison P, Richardson L, Gill AJ. Preva- primary sites as a possible prognostic factor of cutaneous angio- lence of PD-L1 expression in matched recurrent and/or metastatic sarcoma. Int J Cancer. 2014;134:2393–402. sarcoma samples and in a range of selected sarcomas subtypes. 13. Honda Y, Otsuka A, Ono S, Yamamoto Y, Seidel JA, Morita PLoS One. 2020;15:e0222551. S, Hirata M, Kataoka TR, Takenouchi T, Fujii K, Kanekura T, 26. Lee JB, Ahn BC, Kim SH, Lee YH, Han JW, Jeon MK, Kim SH, Okubo Y, Takahashi K, Yanagi T, Hoshina D, Hata H, Abe R, Kim HS. Prognostic implications of PD-L1 expression in patients Fujimura T, Funakoshi T, Yoshino K, Masuzawa M, Amoh Y, with angiosarcoma. Future Sci OA. 2021;7(5):FSO691. Tanaka R, Fujisawa Y, Honda T, Kabashima K. Infiltration of 27. Florou V, Rosenberg AE, Wieder E, Komanduri KV, Kolonias D, PD-1-positive cells in combination with tumor site PD-L1 expres- Uduman M, Castle JC, Buell JS, Trent JC, Wilky BA. Angiosar- sion is a positive prognostic factor in cutaneous angiosarcoma. coma patients treated with immune checkpoint inhibitors: a case Oncoimmunology. 2017;6:e1253657. series of seven patients from a single institution. J Immunother 14. Shimizu A, Kaira K, Okubo Y, Utsumi D, Yasuda M, Asao T, Cancer. 2019;7:213. Nishiyama M, Takahashi K, Ishikawa O. Positive PD-L1 expres- 28. Sindhu S, Gimber LH, Cranmer L, McBride A, Kraft AS. Angio- sion predicts worse outcome in cutaneous angiosarcoma. J Glob sarcoma treated successfully with anti-PD-1 therapy - a case Oncol. 2016;3(4):360–9. report. J Immunother Cancer. 2017;5:58. 15. Bagaria SP, Chang YH, Gray RJ, Ashman JB, Attia S, Wasif N. 29. Xu F, Zheng J, Fu M, Zhou H. Antiprogrammed cell death pro- Improving long-term outcomes for patients with extra-abdom- tein 1 immunotherapy for angiosarcoma with high programmed inal soft tissue sarcoma regionalization to high-volume cent- death-ligand 1 expression: a case report. Immunotherapy. ers, improved compliance with guidelines or both? Sarcoma. 2020;12:771–6. 2018;2018:8141056. 30. Hamacher R, Kampfe D, Reuter-Jessen K, Pottgen C, Podleska 16. Botti G, Scognamiglio G, Marra L, Pizzolorusso A, Di Bonito LE, Farzaliyev F, Steinau HU, Schuler M, Schildhaus HU, Bauer M, De Cecio R, Cantile M, De Chiara A. Programmed death S. dramatic response of a PD-L1-positive advanced angiosarcoma ligand 1 (PD-L1) expression in primary angiosarcoma. J Cancer. of the scalp to pembrolizumab. Jco Precis Oncol. 2018;2:1–7. 2017;8:3166–72. 31. Momen S, Fassihi H, Davies HR, Nikolaou C, Degasperi A, 17. Googe PB, Flores K, Jenkins F, Merritt B, Moschos SJ, Grilley- Stefanato CM, Dias JML, Dasgupta D, Craythorne E, Sarkany R, Olson JE. Immune checkpoint markers in superficial angiosarco - Papa S, Nik-Zainal S. Dramatic response of metastatic cutaneous mas: PD-L1, PD-1, CD8, LAG-3, and tumor-infiltrating lympho - angiosarcoma to an immune checkpoint inhibitor in a patient with cytes. Am J Dermatopathol. 2020. xeroderma pigmentosum: whole-genome sequencing aids treat- 18. D’Angelo SP, Shoushtari AN, Agaram NP, Kuk D, Qin LX, ment decision in end-stage disease. Csh Mol Case Stud. 2019;5. Carvajal RD, Dickson MA, Gounder M, Keohan ML, Schwartz 32. Painter CA, Jain E, Tomson BN, Dunphy M, Stoddard RE, GK, Tap WD. Prevalence of tumor-infiltrating lymphocytes and Thomas BS, Damon AL, Shah S, Kim D, Gomez Tejeda Zanudo PD-L1 expression in the soft tissue sarcoma microenvironment. J, Hornick JL, Chen YL, Merriam P, Raut CP, Demetri GD, Van Hum Pathol. 2015;46(3):357–65. Tine BA, Lander ES, Golub TR, Wagle N. The Angiosarcoma 1 3 268 Immunologic Research (2022) 70:256–268 Project: enabling genomic and clinical discoveries in a rare cancer inhibitor monotherapy in sarcoma patients. J Immunother Cancer. through patient-partnered research. Nat Med. 2020;26:181–7. 2021;9. 33. Weidema ME, van de Geer E, Koelsche C, Desar IME, Kemme- 41. Petitprez F, Meylan M, de Reynies A, Sautes-Fridman C, Fridman ren P, Hillebrandt-Roeffen MHS, Ho VKY, van der Graaf WTA, WH. The Tumor microenvironment in the response to immune Versleijen-Jonkers YMH, von Deimling A, Flucke UE, PGroup. checkpoint blockade therapies. Front Immunol. 2020;11:784. DNA methylation profiling identifies distinct clusters in angiosar - 42. Meyers DE, Banerji S. Biomarkers of immune checkpoint inhibi- comas. Clin Cancer Res. 2020;26(93):100. tor efficacy in cancer. Curr Oncol. 2020;27:S106–14. 34. Casparie M, Tiebosch AT, Burger G, Blauwgeers H, van de Pol A, 43. Petitprez F, de Reynies A, Keung EZ, Chen TW, Sun CM, Calde- van Krieken JH, Meijer GA. Pathology databanking and biobank- raro J, Jeng YM, Hsiao LP, Lacroix L, Bougouin A, Moreira M, ing in The Netherlands, a central role for PALGA, the nationwide Lacroix G, Natario I, Adam J, Lucchesi C, Laizet YH, Toulmonde histopathology and cytopathology data network and archive. Cell M, Burgess MA, Bolejack V, Reinke D, Wani KM, Wang WL, Oncol. 2007;29:19–24. Lazar AJ, Roland CL, Wargo JA, Italiano A, Sautes-Fridman C, 35. Weidema ME, Flucke UE, van der Graaf WTA, Ho VKY, Hille- Tawbi HA, Fridman WH. B cells are associated with survival and brandt-Roeffen MHS, Dutch Nationwide N, Registry of H, Cyto - immunotherapy response in sarcoma. Nature. 2020;577:556–60. pathology G, Versleijen-Jonkers YMH, Husson O, Desar IME. 44. Li J, Wei Q, Wu X, Sima J, Xu Q, Wu M, Wang F, Mou H, Hu H, Prognostic Factors in a Large nationwide cohort of histologi- Zhao J, Li D, Hu J, Zhang L, Zhu X, Chen L, Luo C, Yan J, He J, cally confirmed primary and secondary angiosarcomas. Cancers Ma Y, Shao Y, Wu W, Ying J. Integrative clinical and molecular (Basel). 2019;11. analysis of advanced biliary tract cancers on immune checkpoint 36. van Erp AEM, Versleijen-Jonkers YMH, Hillebrandt-Roeffen blockade reveals potential markers of response. Clin Transl Med. MHS, van Houdt L, Gorris MAJ, van Dam LS, Mentzel T, Wei- 2020;10:e118. dema ME, Savci-Heijink CD, Desar IME, Merks HHM, van 45. Borghaei H, Paz-Ares L, Horn L, Spigel DR, Steins M, Ready Noesel MM, Shipley J, van der Graaf WTA, Flucke UE, Meyer- NE, Chow LQ, Vokes EE, Felip E, Holgado E, Barlesi F, Kohl- Wentrup FAG. Expression and clinical association of programmed haufl M, Arrieta O, Burgio MA, Fayette J, Lena H, Poddubs - cell death-1, programmed death-ligand-1 and CD8(+) lympho- kaya E, Gerber DE, Gettinger SN, Rudin CM, Rizvi N, Crino cytes in primary sarcomas is subtype dependent. Oncotarget. L, Blumenschein GR Jr, Antonia SJ, Dorange C, Harbison CT, 2017;8:71371–84. Graf Finckenstein F, Brahmer JR. Nivolumab versus Docetaxel 37. Cooper WA, Tran T, Vilain RE, Madore J, Selinger CI, Kohonen- in Advanced nonsquamous non-small-cell lung cancer. N Engl J Corish M, Yip P, Yu B, O’Toole SA, McCaughan BC, Yearley JH, Med. 2015;373:1627–39. Horvath LG, Kao S, Boyer M, Scolyer RA. PD-L1 expression is 46. Garon EB, Rizvi NA, Hui R, Leighl N, Balmanoukian AS, Eder a favorable prognostic factor in early stage non-small cell carci- JP, Patnaik A, Aggarwal C, Gubens M, Horn L, Carcereny E, noma. Lung cancer (Amsterdam, Netherlands). 2015;89:181–8. Ahn MJ, Felip E, Lee JS, Hellmann MD, Hamid O, Goldman JW, 38. Schalper KA, Velcheti V, Carvajal D, Wimberly H, Brown J, Pusz- Soria JC, Dolled-Filhart M, Rutledge RZ, Zhang J, Lunceford JK, tai L, Rimm DL. In situ tumor PD-L1 mRNA expression is associ- Rangwala R, Lubiniecki GM, Roach C, Emancipator K, Gandhi L, ated with increased TILs and better outcome in breast carcinomas. Investigators K. Pembrolizumab for the treatment of non-small- Clin Cancer Res. 2014;20:2773–82. cell lung cancer. N Engl J Med. 2015;372:2018–28. 39. Chan JY, Lim JQ, Yeong J, Ravi V, Guan P, Boot A, Tay TKY, 47. Zajac M, Ye J, Mukhopadhyay P, Jin X, Ben Y, Antal J, Gupta Selvarajan S, Md Nasir ND, Loh JH, Ong CK, Huang D, Tan J, Li AK, Rebelatto MC, Williams JA, Walker J. Optimal PD-L1-high Z, Ng CC, Tan TT, Masuzawa M, Sung KW, Farid M, Quek RHH, cutoff for association with overall survival in patients with urothe - Tan NC, Teo MCC, Rozen SG, Tan P, Futreal A, Teh BT, Soo KC. lial cancer treated with durvalumab monotherapy. PLoS One. Multiomic analysis and immunoprofiling reveal distinct subtypes 2020;15:e0231936. of human angiosarcoma. J Clin Invest. 2020;130(11):5833–46. 48. Schildhaus HU. Predictive value of PD-L1 diagnostics. Pathologe. 40. Starzer AM, Berghoff AS, Hamacher R, Tomasich E, Feldmann 2018;39:498–519. K, Hatziioannou T, Traint S, Lamm W, Noebauer-Huhmann IM, Furtner J, Mullauer L, Amann G, Bauer S, Schildhaus HU, Publisher's note Springer Nature remains neutral with regard to Preusser M, Heller G, Brodowicz T. Tumor DNA methylation jurisdictional claims in published maps and institutional affiliations. profiles correlate with response to anti-PD-1 immune checkpoint 1 3

Journal

Immunologic ResearchSpringer Journals

Published: Apr 1, 2022

Keywords: Angiosarcoma; Subgroups; Programmed cell death 1; Programmed death-ligand 1 (PD-L1); Immune checkpoint inhibition

There are no references for this article.