If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address for reprints: Marc de Perrot, MD, MSc, FRCSC, University Health Network, University of Toronto, 9N-961, 200 Elizabeth St, Toronto, Ontario M5G 2C4, Canada.
Division of Thoracic Surgery, University Health Network, University of Toronto, Toronto, Ontario, CanadaLatner Thoracic Surgery Laboratories, University Health Network, University of Toronto, Toronto, Ontario, CanadaDepartment of Immunology, University of Toronto, Toronto, Ontario, Canada
Division of Thoracic Surgery, University Health Network, University of Toronto, Toronto, Ontario, CanadaLatner Thoracic Surgery Laboratories, University Health Network, University of Toronto, Toronto, Ontario, Canada
Cytotoxic CD8+ tumor infiltrating lymphocytes (TILs) can contribute to the benefit of hypofractionated radiation, but programmed cell death pathways (programmed cell death 1 and programmed cell death ligand 1 [PD-1/PD-L1]) may provide a mechanism of tumor immune escape. We therefore reviewed the influence of PD-1/PD-L1 and CD8+ TILs on survival after accelerated hypofractionated hemithoracic radiation followed by extrapleural pneumonectomy for malignant pleural mesothelioma (MPM).
Methods
Sixty-nine consecutive patients undergoing the protocol of Surgery for Mesothelioma after Radiation Therapy (SMART) between November 2008 and February 2016 were analyzed for the presence of PD-L1 on tumor cells, PD-1 on inflammatory cells, and CD8+ TILs. Comparison was made with a cohort of patients undergoing extrapleural pneumonectomy after induction chemotherapy (n = 14) and no induction (n = 2) between March 2005 and October 2008. PD-L1 expression on tumor cells ≥1% was considered positive. CD8+ TILs and PD-1 expression were scored as a percentage of positive cells.
Results
PD-L1 was negative in 75% of MPM after completion of SMART. CD8+ TILs ranged between 0.24% and 8.47% (median 2%). CD8+ TILs ≥2% was associated with significantly better survival in epithelioid MPM (median survival 3.7 years vs 2.3 years in CD8+ TILs <2%; P = .02). PD-L1 positivity was associated with worse survival in biphasic MPM (median survival, 0.4 years vs 1.5 years in biphasic PD-L1 negative tumors; P = .07) after SMART. Multivariate analysis demonstrated that epithelioid MPM, nodal disease, and CD8+ TILs were independent predictors of survival after SMART.
Conclusions
The influence of tumor microenvironment on survival differs between epithelioid and nonepithelioid MPM. CD8+ TILs is an independent factor associated with better survival in epithelioid MPM treated with SMART.
A, Representative high power field of the CD8-stained tumor tissue with 50 micron scale bar. B, Nucleus detection (black outline and blue or yellow objects). C, The same region with positive cells above the immunohistochemistry threshold (brown), relative to negative cells (white). The high-power field is 250 microns in width and height.
Hypofractionated accelerated radiation followed by surgery can have a beneficial influence on the immune system in epithelioid mesothelioma. This observation supports the potential role of this approach as a platform for immunotherapy in mesothelioma. The choice of immunotherapy may need to be adjusted according to the type of mesothelioma.
Malignant pleural mesothelioma (MPM) remains associated with dismal prognosis despite aggressive therapy with chemotherapy, surgery, and radiation therapy. Immunotherapy provides new potential opportunities, but results with checkpoint inhibitors have had limited success so far.
Tremelimumab as second-line or third-line treatment in relapsed malignant mesothelioma (DETERMINE): a multicentre, international, randomised, double-blind, placebo-controlled phase 2b trial.
Hence, better understanding of the mesothelioma tumor immune microenvironment is critical to select appropriate immunotherapeutic targets and define potential biomarkers of response.
Evidence suggests that the tumor microenvironment (TME) profoundly differs between epithelioid and nonepithelioid mesotheliomas.
Hence, the subtype of mesothelioma may influence the selection of immunotherapeutic targets and biomarkers. These differences may also explain the discrepancy in the role of CD8+ T cells in the TME. Some studies demonstrated that high CD8+ T cells in the mesothelioma TME was associated with better prognosis, whereas others have shown that it was a marker of worse outcome.
Malignant pleural mesothelioma immune microenvironment and checkpoint expression: correlation with clinical-pathological features and intratumor heterogeneity over time.
Recent evidence have shown that hypofractionated radiation can prime the immune system against the tumor and contribute to the benefit of radiation therapy.
Tumor-infiltrating lymphocytes (TILs) and, in particular, cytotoxic CD8+ T cells, can contribute to the benefit of hypofractionated radiation, but programmed cell death pathways (ie, programmed cell death 1/programmed cell death ligand 1 [PD-1/PD-L1]) may provide a mechanism of tumor immune escape.
We therefore reviewed the influence of PD-1/PD-L1 and CD8+ TILs on survival after hypofractionated accelerated hemithoracic radiation followed by extrapleural pneumonectomy (EPP) for MPM in our cohort of patients undergoing Surgery for Mesothelioma After Radiation Therapy (SMART).
Formalin-fixed, paraffin-embedded tissues were obtained from 80 consecutive patients undergoing SMART in our institution between November 2008 and February 2016. The SMART approach was previously described in detail.
Briefly, patients with histology-proven mesothelioma undergo an accelerated course of hemithoracic hypofractionated radiation of 25 to 30 Gy in 5 fractions followed by EPP within a week from the end of radiation. Preoperative staging included computed tomography scan of the chest and abdomen, positron-emission computed tomography, brain imaging, and bronchoscopy with endobronchial ultrasound fine-needle aspiration biopsy of the mediastinal and hilar lymph nodes since December 2013. The SMART protocol has become our preferred approach with increasing experience for surgically resectable MPM. All tissues were collected under an institutional ethical board-approved protocol (No. 19-5122). Histopathologic diagnosis was established by our institution's thoracic pathologists (MC and MST). The diagnosis of biphasic mesothelioma was established by the presence of at least 10% spindle cells within the tumor specimen. Patients who died from complications without recurrence within 90 days of surgery (n = 3) were excluded from the study because the primary goal of the study was to assess the prognostic influence of CD8+ TILs and PD-1/PD-L1 pathway. Patients without adequately preserved tumor specimens were also excluded from the study (n = 4). Patients undergoing preoperative radiation and chemotherapy were excluded from the study because preoperative chemotherapy was not part of the SMART approach and these patients were typically sent for a surgical assessment after completion of chemotherapy (n = 4). The levels of CD8+ TILs as well as PD-1 and PD-L1 expression in the SMART cohort was then compared with a historical cohort of patients undergoing EPP after either induction chemotherapy (n = 14) or without any induction therapy (n = 2) between March 2005 and October 2008 (Figure 1). Patients' characteristics are summarized in Table 1. All patients were followed until death or for at least 5 years after their initial treatment. On last follow-up in October 2018, 69 patients (81%) had died, 9 were alive with disease, and 7 were alive without disease.
Figure 1Consolidated standards of reporting trials diagram. SMART, Surgery for Mesothelioma after Radiation Therapy; EPP, extrapleural pneumonectomy; TIL, tumor-infiltrating lymphocyte; PD-1, programmed cell death 1; PD-L1, programmed cell death ligand 1.
Formalin-fixed paraffin-embedded blocks of selected patients were cut in serial sections of 4 μm depth and stained with hematoxylin and eosin, CD8, PD-1, and PD-L1. The endogenous peroxidase activity was blocked with 3% hydrogen peroxide. Antigen retrieval or unmasking procedure was applied by Tris-EDTA (Sigma-Aldrich, Oakville, Ontario, Canada) (pH9.0) for CD8 and CC1 Ventana Retrieval Buffer (Roche, Mississauga, Ontario, Canada) for PD-1 and PD-L1, respectively. Serum block was applied according to the instructions of ImmPress species-specific HRP kit (Vector Laboratories Inc, Burlingame, Calif). Primary antibodies against CD8 (Clone: 4B11VPC-324; Leica Biosystems Inc, Concord, Ontario, Canada), PD-1 (Clone: NAT105; Abcam Inc, Toronto, Ontario, Canada), and PD-L1 (Clone: 28.8; Abcam Inc, Toronto, Ontario, Canada) were used 1/1000 1 hour for CD8, and 1/200 overnight for both PD-1 and PD-L1. The MACH 1 Universal Polymer Detection System containing 4 reagents (Intermedico catalog No. BC-M4U534L; Markham, Ontario, Canada) was applied per kit instructions, to provide superior staining and a reduced protocol time compared with the conventional system. DAB kit (DAKO catalog No. K3468; Mississauga, Ontario, Canada) was used to develop color. Sections were mounted with MM 24 Leica mounting medium (catalog No. 3801120) after counterstaining lightly with Mayer's hematoxylin (Sigma-Aldrich).
PD-L1 Analysis
PD-L1 scoring on tumor cells was performed by 2 thoracic pathologists (MC and MST). PD-L1 expression on tumor infiltrating cells were not counted. Tumors with positive PD-L1 expression on 1% or more of tumor cells were considered positive, regardless of intensity. Positive staining was then divided into 1% to 4% (low), 5% to 49% (intermediate), and ≥50% positivity (strong) according to previous publications.
Malignant pleural mesothelioma immune microenvironment and checkpoint expression: correlation with clinical-pathological features and intratumor heterogeneity over time.
In case of discrepancy, the highest PD-L1 expression was reported.
CD8 and PD-1 Analysis
Mesothelioma tissues stained for CD8 and PD-1 were first annotated by a pathologist to highlight the area of interest. Quantification of positive cell staining was performed on whole slide scanned with Definiens Tissuestudio 4.0 software (Definiens Inc, Carlsbad, Calif) utilizing nucleus detection, cell simulation, and cell classification algorithms following a similar methodology described previously.
Evaluation of optimal biopsy location for assessment of histological activity, transcriptomic and immunohistochemical analyses in patients with active Crohn's disease.
In brief, regions of interest corresponding to mesothelioma tumor were manually annotated by a pathologist to exclude artifacts and nontumoral area. Following transfer of those annotations to each set of stained images, stain separation was used to isolate the brown immunohistochemistry stain from blue hematoxylin counterstain. Within these regions of interest, a nucleus detection algorithm utilized the blue hematoxylin counterstain above a threshold to identify the nuclei of cells, plus a watershed step to break up clusters of nuclei (Figure 2). Cell detection simulated a cytoplasmic/membrane region of a few microns in thickness. Results were summarized as a percentage of positive cells for each marker, measured by counting the number of positive cells and dividing by the total number of cells detected in the region of interest. A subset of high power fields were visually scored by 2 investigators (LW and TM) for manual validation of percent cell counts.
Figure 2A, Representative high power field of the CD8 stained tumor tissue. A 50 micron scale bar is included. B, Nucleus detection (black outline and blue or yellow objects). C, The same region with positive cells above the immunohistochemistry threshold (brown), relative to negative cells (white). The high power field is 250 microns in width and height.
Demographic characteristics and results were reported as mean ± standard deviation or as median and range. Categorical variables were compared by χ2 analysis and continuous variables by Student t test. One-way analysis of variance was used for multiple comparisons. Survival was estimated using the Kaplan-Meier method. Survival was calculated from the starting day of treatment; that is, radiation, chemotherapy, or surgery in the absence of preoperative therapy up to the date of death or last follow-up for patients still alive. Differences in survival were tested for significance using the log-rank test. The median values of CD8+ TILs and PD-1 expression were used as a cutoff to calculate survival. Univariate and multivariate Cox regression analysis was used to determine factors associated with survival. Correlations were evaluated by linear regression analysis. GraphPad Prism 6 statistical software (La Jolla, Calif) and Statview V (Abacus Concept, Berkeley, Calif) were used for all analyses.
Results
Among the cohort of patients undergoing the SMART approach, PD-L1 expression was negative (<1%) in 75% of patients. In PD-L1+ tumors, PD-L1 expression was low in 20% of patients and intermediate to strong in the remaining 3 patients. Overall, 90% of slides were consistently positive or negative between both pathologists (κ = 0.66). In total, 35 out of 47 patients with epithelioid mesothelioma (74.5%) and 17 out of 22 patients with nonepithelioid mesothelioma (77.2%) had negative PD-L1 expression. PD-L1 expression was not influenced by nodal disease with negative expression in 71% of patients with N0 disease and 78% of patients with positive nodes (P = .5). Pathological stage did not influence PD-L1 expression with 25% PD-L1 positivity in stage I or II, 27% in stage III, and 24% in stage IV (P = .9).
CD8+ TILs ranged between 0.24% and 8.5% (median, 2%) and PD-1 expression between 0% and 7.5% (median, 0.3%). The proportions of CD8+ TILs was similar between epithelioid and nonepithelioid mesothelioma (2.3% ± 1.9% vs 2.4% ± 1.5%, respectively; P = .8) and between patients with positive nodes (N1 and/or N2, N+) and negative nodes (2.3% ± 1.5% vs 2.4% ± 1.9%, respectively; P = .8). PD-1 expression was also similar between epithelioid and nonepithelioid mesothelioma (0.8% ± 1.3% vs 0.7% ± 0.8%, respectively; P = .7). However, patients with positive nodes had a lower level of PD-1 expression compared with node-negative patients (0.5% ± 0.7% vs 1.2% ± 1.6%, respectively; P = .02).
Factors potentially affecting survival in the cohort of patients undergoing SMART included histology, proportion of CD8+ TILs, nodal disease, and gender (Table 2). In multivariate analysis, histology, nodal disease, and CD8+ TILs remained independent predictors of survival (Table 2). The independent benefit of CD8+ TILs on survival was seen whether CD8+ TILs was analyzed as a continuous variable or a dichotomized variable based on the median values. Survival was best in patients with epithelioid disease and CD8+ TILs >2% with a median survival of 45 months and a 5-year survival of 42% (Figure 3). PD-L1 positivity was associated with worse survival in biphasic disease with a median survival of 5 months (Figure 3). PD-L1 positivity in epithelioid MPM was associated with a better survival than PD-L1 negative epithelioid MPM with a median survival of 66 months versus 32 months, respectively (P = .06). CD8+ TILs positively correlated with survival in epithelioid node positive MPM (Figure 4) with a median survival reaching 30 months in CD8+ TILs >2% compared with 17 months in CD8+ TILs <2% (P = .02).
Table 2Univariate and multivariate analysis for survival in the Surgery For Mesothelioma after Radiation Therapy cohort
Patient characteristic
Univariate analysis
Multivariate analysis
n
Median survival (mo)
P value
Hazard ratio (95% Confidence interval)
P value
Positive N1 or N2 lymph nodes
.05
1.92 (1.05-3.51)
.03
N0
26
43
N+
42
19
Epithelioid histology
.0007
0.3 (0.16-0.59)
.0004
Epithelioid
49
34
Biphasic
19
14
Male gender
.07
1.79 (0.79-4.06)
.2
Male
55
23
Female
13
51
PD-L1– positive cancer cells
.2
0.94 (0.43-2.06)
.9
Positive
17
27
Negative
51
24
PD-1 cells >0.3%
.7
0.96 (0.71-1.32)
.8
>0.3%
32
24
<0.3%
36
28
CD8+ TILs >2%
.03
0.47 (0.25-0.89)
.02
>2%
33
36
<2%
35
20
PD-L1, Programmed cell death ligand 1; PD-1, programmed cell death 1; TILs, tumor-infiltrating lymphocytes.
Figure 3A, Survival after the Surgery for Mesothelioma after Radiation Therapy (SMART) approach according to histology and CD8+ TILs. Survival after SMART was influenced by histology and CD8+ tumor-infiltrating lymphocytes (TILs). The survival was better in patients with epithelioid mesothelioma and CD8+ TILs ≥2% than in patients with epithelioid mesothelioma and CD8+ TILs <2% (P = .02). CD8+ TILs did not influence survival in biphasic tumors. B, Survival after the SMART approach according to histology and programmed cell death ligand 1 (PD-L1) expression. Survival after SMART was influenced by histology and PD-L1 expression. PD-L1 expression had a different influence in epithelioid and biphasic tumors. The survival was better in patients with epithelioid PD-L1 positive tumors compared with epithelioid PD-L1 negative tumors (P = .06) and worse in biphasic PD-L1 positive tumors compared with biphasic PD-L1 negative tumors (P = .07). The confidence limits as well as the number of patients at risk are available in Table E1.
Figure 4Correlation between percentage of CD8+ tumor-infiltrating lymphocytes (TILs), lymph node positivity, and survival in patients undergoing the Surgery For Mesothelioma after Radiation Therapy (SMART) approach. The correlation was positive between CD8+ TILs and survival in N+ patients (R2, 0.24; P = .006).
The cohort of patients undergoing SMART was then compared with patients undergoing either induction chemotherapy before surgery or no induction therapy. CD8+ TILs were lower after SMART than after induction chemotherapy or no induction, but the proportion of PD-1 expression was similar among all 3 groups of patients (Figure 5). The median survival reached 34.4 months in patients with epithelioid MPM treated with SMART compared with 21.6 months in patients with epithelioid MPM treated with induction chemotherapy and EPP (P = .5). When analyzing all patients together, CD8+ TILs had a beneficial influence on survival in epithelioid mesothelioma after SMART (continuous variable, P = .03; 95% confidence interval [CI], 0.67-0.98), but not after induction chemotherapy (continuous variable, P = .2; 95% CI, 0.85-1.05). When analyzing nonepithelioid mesothelioma separately, PD-L1 expression was the only independent predictor of survival (P = .003; 95% CI, 2.2-53.3) with a median survival of 6 months in PD-L1+ tumors compared with 18 months in PD-L1– tumors.
Figure 5A, Percentage of CD8+ tumor-infiltrating lymphocytes (TILs) according to the preoperative therapy. The percentage of CD8+ TILs was significantly lower after Surgery For Mesothelioma after Radiation Therapy (SMART) compared with preoperative chemotherapy and to sarcomatoid mesothelioma without induction therapy. The highest level of CD8+ TILs was observed in sarcomatoid tumors without induction therapy. B, Percentage of programmed cell death 1 (PD-1) expression according to the preoperative therapy. The level of PD-1 expression was similar across all groups with no significant difference between groups. *P < .05. **P < .01.
This study confirms the importance of identifying the histologic subtype of mesothelioma when analyzing the TME. We observed that CD8+ TILs were an independent predictor of survival in epithelioid MPM after the SMART approach despite the presence of other important prognostic factors such as histology and nodal disease.
CD8+ TILs had no influence on survival in nonepithelioid MPM.
Different immune microenvironment in epithelioid and nonepithelioid mesothelioma can explain the influence of different immune factors on survival. Epithelioid mesothelioma typically have an immune-activated TME with greater proportion of plasmacytoid dendritic cells, CD20+ cells, CD4+ helper T cells, and exhausted CD8+ TILs, whereas nonepithelioid mesothelioma have an immune privileged TME with large proportions of macrophages, regulatory T cells, mesothelioma stem cells, and neutrophils.
Although the proportion of CD8+ T cells can be high in both TMEs, CD8+ TILs are typically polyclonal and less cytotoxic in immune privilege microenvironment, and more oligoclonal and cytotoxic in immune activated microenvironment.
Hence, the reactivity (ie, quality) of CD8+ T cells against specific tumor antigens appears more important than the number (ie, quantity) of CD8+ TILs. This observation can explain the conflicting results in the prognosis of CD8+ T cells seen in mesothelioma. Some studies performed in surgical patients with predominantly or exclusively epithelioid mesothelioma have shown that the presence of CD8+ TILs were associated with better outcome.
In contrast, several recent studies performed in nonsurgical patients with large proportions of nonepithelioid mesothelioma demonstrated that high proportion of CD8+ TILs was a marker of poor outcome.
Malignant pleural mesothelioma immune microenvironment and checkpoint expression: correlation with clinical-pathological features and intratumor heterogeneity over time.
This high proportion of CD8+ TILs was associated with high level of PD-L1 expression suggesting that an immune-privileged TME was prominent in these patients and likely different than surgical patients with epithelioid disease.
Malignant pleural mesothelioma immune microenvironment and checkpoint expression: correlation with clinical-pathological features and intratumor heterogeneity over time.
The results from SMART shows the importance of CD8+ T cells as a positive prognostic marker in our cohort of patients with epithelioid mesothelioma. Although preoperative radiation was associated with a drop in CD8+ TILs at the time of surgery compared with patients undergoing induction chemotherapy, the association between CD8+ TILs and survival in epithelioid mesothelioma suggests that the CD8+ TILs identified within the specimen could be active against specific epitopes released by the tumor and more effective in controlling the disease. We are thus planning additional work to sequence the T cell receptor of these CD8+ T cells to determine their clonality.
We and others have shown in preclinical animal models that a short course of nonablative hypofractionated radiation can activate the immune system against the tumor by releasing tumor associated antigens, danger signals, and type I interferon (IFN) in the TME.
Increasing evidence suggests that the proinflammatory microenvironment generated by IFN-γ can also lead to the upregulation of regulatory T cells and the upregulation of PD-L1.
Hence, although necessary for the proinflammatory response generated by radiation, IFN-γ may also limit the benefit of radiation and not be an adequate marker of response to radiation. A schematic of the immune activation process generated by an accelerated course of nonablative radiation is presented in Figure 6.
Figure 6An accelerated course of hypofractionated radiation is directed to the tumor, leading to an immunogenic cell death with release of double stranded DNA (dsDNA), tumor associated antigens (TAA), and type I interferon (IFN-α and IFN-β) that leads to an activation of dendritic cells in the draining lymph nodes and, consequently, the priming of CD8+ T cells (CD8) against the tumor. Tumor cells then upregulate programmed cell death 1 (PD-L1) as a mechanism of protection against antitumoral CD8+ T cells.
We observed that low level of PD-L1 expression (1%-4%) in epithelioid mesothelioma in the SMART cohort was associated with better survival compared with PD-L1– epithelioid tumor. This effect could be due to some of the variability in scoring PD-L1 when the expression is low.
Because this study is a snap shot of the immune system in the TME 1 week after the end of radiation, we do not know how long the proinflammatory milieu generated by the short course of radiation lasts. However, continuous IFN-γ release can contribute to an epithelial-to-mesenchymal transition, which may explain some of the aggressive recurrences that we occasionally see after SMART even a few years after the initial treatment.
The optimal PD-L1 cutoff to define positivity is still unknown in mesothelioma. A high cutoff could help identify the truly positive tumors with PD-L1 expression resulting from inappropriately activated signaling pathway in tumor cells.
High PD-L1 expression provides a permissive immune microenvironment by tolerizing T cells and is often associated with prominent inflammation and high levels of CD8+ TILs.
However, as mentioned previously, these CD8+ TILs are often polyclonal with limited cytotoxicity and, thus, less specific in their targeting and not as effective in controlling the disease. Immune checkpoint blockade targeting the PD1/PD-L1 pathway can unleash a powerful immune response in some of these patients suggesting that the activity of CD8+ cells can occasionally predominantly be restraint by the PD1/PD-L1 pathway.
The influence of CD8+ T cells on survival in epithelioid mesothelioma after SMART support the concept that this short course of hypofractionated hemithoracic radiation could be used as a platform for immunotherapy. The optimal immunotherapeutic targets are unknown and remain to be determined. Immune checkpoints inhibitors in mesothelioma have not demonstrated the similar successes seen in melanoma and non–small cell lung cancers and, thus, other approaches with costimulatory molecules and cytokines are warranted. The Cancer Genome Atlas report demonstrated that V-domain Ig suppressor of T cell activation was highly expressed in epithelioid mesothelioma, whereas biphasic and sarcomatoid mesothelioma predominantly expressed CD134 (OX40)–OX40 ligand and B7-H3 CD276 (B7-H3) in association with markers of epithelial-mesenchymal transition.
V-domain Ig suppressor of T cell activation, OX40, and B7-H3 could thus be potential targets for immunotherapy.
This study presents several limitations related to the retrospective nature of the study, the absence of preoperative tissue to assess the TME before starting the radiation, and the specific focus on the PD1/PD-L1 pathway and CD8+ T cells. Both PD-L1 and CD8+ TILs are important prognostic markers associated with controversial findings in the literature. Hence, our findings do shed some light to explain these discordant findings.
We are currently in the process of implementing a new clinical trial to refine the hypofractionated accelerated dose of radiation given before surgery in MPM. Based on our preclinical models and data from the literature, we are planning to use 3 fractions of radiation instead of 5, and to increase the hypofractionated boost dose of radiation to the gross tumor volume to at least 21 Gy to better potentiate the abscopal effect.
The dose of radiation to the whole hemithorax will be decreased to preserve the underlying lung function while still providing activation of the immune system. This protocol will include specific tissue collections to determine the optimal dose of radiation to generate immune activation and define potential targets for immunotherapy.
Conclusions
Our study demonstrates the importance of understanding the TME in predicting outcome in mesothelioma. The prognostic markers are different between epithelioid and nonepithelioid mesothelioma and therefore the subtype of mesothelioma should be taken into consideration when immune biomarkers are being developed for mesothelioma. CD8+ TILs are an independent prognostic marker for survival after SMART, supporting the potential influence of this hypofractionated accelerated radiation course on the immune system. Nonepithelioid mesothelioma can express intermediate to high levels of PD-L1 associated with poor outcome. In contrast, epithelioid mesothelioma can express low levels of PD-L1 associated with good outcome. This observation suggests that higher PD-L1 expression threshold may be better to define positivity.
Authors have nothing to disclose with regard to commercial support.
Appendix
Table E1The confidence limits as well as the number of patients at risk in the Surgery for Mesothelioma after Radiation Therapy (SMART) approach groups described in Figure 3
Variable
Epithelioid mesothelioma
Epithelioid mesothelioma
Biphasic tumor
Biphasic tumor
CD8 >2%
CD8 <2%
CD8 >2%
CD8 <2%
Standard error of the mean
Year
0
100
100
100
100
1
100
79 ± 8
56 ± 17
60 ± 15
2
77 ± 9
50 ± 10
22 ± 14
20 ± 13
3
64 ± 10
38 ± 10
22 ± 14
0
4
48 ± 11
28 ± 9
0
5
42 ± 11
23 ± 9
6
42 ± 11
0
No. of patients at risk
Year
0
22
24
9
10
1
22
20
6
6
2
17
12
2
2
3
14
9
1
4
8
6
1
5
3
4
6
1
1
PD-L1–
PD-L1+
PD-L1–
PD-L1+
Standard error of the mean
Year
0
100
100
100
100
1
83 ± 6
100
71 ± 11
20 ± 18
2
60 ± 8
67 ± 14
29 ± 11
20 ± 18
3
46 ± 8
58 ± 14
18 ± 9
0
4
30 ± 8
58 ± 14
18 ± 9
5
21 ± 8
58 ± 14
9 ± 8
6
11 ± 8
29 ± 16
No. of patients at risk
Year
0
35
12
17
5
1
29
12
12
2
2
21
8
5
1
3
16
7
2
4
8
6
2
5
2
5
6
1
1
PD-L1, Programmed cell death ligand 1; PD-1, programmed cell death 1.
Tremelimumab as second-line or third-line treatment in relapsed malignant mesothelioma (DETERMINE): a multicentre, international, randomised, double-blind, placebo-controlled phase 2b trial.
Malignant pleural mesothelioma immune microenvironment and checkpoint expression: correlation with clinical-pathological features and intratumor heterogeneity over time.
Evaluation of optimal biopsy location for assessment of histological activity, transcriptomic and immunohistochemical analyses in patients with active Crohn's disease.
For treating malignant pleural mesothelioma, multimodality therapies consisting of chemotherapy, surgery, and radiation therapy are used worldwide for achieving better survival. As a surgical modality, pleurectomy or decortication has also been widely performed, along with extrapleural pneumonectomy. Furthermore, similar to the treatment of primary lung cancer, immunotherapy has recently been introduced for the treatment of malignant pleural mesothelioma. Despite these therapeutic advances, however, the prognosis of malignant pleural mesothelioma remains poor.
All thoracic surgeons by this time are aware of the phenomenal results in non–small cell lung cancer when immunotherapy is combined with chemotherapy.1 These effects occur at the level of the tumor microenvironment, heterogeneously populated with T cells, dendritic cells, macrophages, as well as granulocytes. Checkpoint inhibition will “take the brakes off” of the inhibitory effect of immune cytotoxicity by modulating the PD-L1/PD-1 axis; however, studies have shown that you can get more “bang for your buck” when you add other therapies to the checkpoint inhibitors.
32514-0&id=fx1.jpg){kind=link}
32514-0&id=fx2.jpg){kind=link}
32514-0&id=gr1.jpg){kind=link}
32514-0&id=gr2.jpg){kind=link}
32514-0&id=gr3.jpg){kind=link}
32514-0&id=gr4.jpg){kind=link}
32514-0&id=gr5.jpg){kind=link}
32514-0&id=gr6.jpg){kind=link}
32514-0&id=fx3.jpg){kind=link}