According to a small study that examined 13 peritoneal samples, PDL1 was expressed in a significant portion of malignant mesothelioma. They found the percentage of tissue microarray samples positive for PDL1 expression (defined as >1% tumor staining) using two FDA-approved immunohistochemistry markers to be significantly higher in MPeM (50-60%), than in pleural mesothelioma (18-22%). Another more recent study supported the finding of high frequency of PDL1 expression in patients with MPeM (13). These results in addition to the available evidence of the complex pro-inflammatory microenvironment present in MPeM provided the foundation and reasoning for investigating immune-based treatment strategies like PDL1 inhibition with ICIs. The findings also supported investigating PDL1 as a potential biomarker for immunotherapy benefit in a subset of MPeM patients in clinical trials (14-16). Interestingly, PDL1 expression is temporally heterogeneous in MPeM. This has been linked to past exposure to cytotoxic chemotherapy as it was shown that patients who received systemic chemotherapy in the past had a significantly lower frequency of PDL1 expression (6% vs. 16%) (17). This highlights the importance of determining the optimal treatment sequence for successful outcomes and should be taken into account when planning future clinical studies. For the time being, there is no consensus on selecting patients with MPeM who would benefit from therapies that target PD-1 using the expression of PDL1.

It has been demonstrated that tumors with high levels of microsatellite instability (MSI-H) are hypermutated and produce many peptides that function as neoantigens, causing a robust immune response that is characterized by a high number of tumor-infiltrating lymphocytes (TILs) (18-21). Previous evidence has shown that MSI-H cancers are responsive to treatment with PD-1 ICIs (22). One example is colorectal cancer, which demonstrated an impressive and durable response to ICIs (18,23-25). Colorectal, endometrial, and gastric adenocarcinomas are the most common cancers in which MSI has been identified (26,27). Unfortunately, the occurrence of MSI-H in mesothelial cancer is extremely rare (28). In the rare event of identifying MSI-H, pembrolizumab, which is a PD-1 inhibitor, was granted approval by the FDA in May 2017 for the treatment of advanced solid tumors, including mesothelioma, that has MSI-H and have progressed following prior treatment, and for which there are no alternative treatment options that are considered to be satisfactory.

Immunotherapy appears to be more effective in cancers with high tumor mutation burden (TMB-H) (defined as 10 or more mutations per megabase), such as melanoma and non-small cell lung carcinoma (29,30). Mesothelioma patients have the lowest prevalence of TMB-H across all solid malignancies (1.2%) (31). Of note, patients with MPeM who have a higher somatic mutational burden, had a significant increase in PDL1 staining. The same finding of a significant increase in PDL1 staining was noted in patients with known deleterious germline mutations (17). Again, pembrolizumab was granted accelerated approval by the FDA in June of 2020 for the treatment of unresectable or metastatic solid tumors with TMB-H in patients who have progressed in response to prior treatment and have no alternative treatment options that are considered to be satisfactory.

The immunological microenvironment of MPeM is little understood. Pleural mesothelioma has taught us that TILs may have an impact on the prognosis of patients with mesothelioma. TILs were reported to have a favorable prognostic effect in pleural mesothelioma in early immunohistochemistry (IHC)-based research (32,33). Conversely, a strong negative correlation was found between the stromal TILs score and survival in a cohort of 329 pleural malignant mesothelioma. Nevertheless, the presence of tertiary lymphoid structures (TLS), which serve as access points for TILs entry and local priming, was independently associated with higher survival when multivariate analysis was performed (34). The evaluation of immune infiltrates in tumors, is gaining importance as a prognostic and potentially predictive biomarker to select patients who have the highest likelihood of responding to immunotherapeutic agents. TILs assessment has not been established as a clinical biomarker yet; however, it has been suggested to include it in the routine histopathological reporting (35-37). TILs have been thought of as a favorable prognostic marker however, some patients with high TILs levels may not have a better prognosis. A possible explanation is that TILs levels measured on hematoxylin and eosin (H&E) may not reflect levels of active anti-tumor TILs reflecting functional inhibition. TILs may be exhausted or rendered inactive by immune checkpoint pathways signaling or by a lack of immune stimulatory pathways which both can be reactivated through immunotherapy (38,39). Alternatively, generic TILs detection by H&E may not account for T regs or other immunosuppressive populations (40-42).

EMT genes are unique potential targets that can be used to halt the progression of cancer and its spread (43). It has been demonstrated that EMT plays an essential role in the process of carcinogenesis, as well as the progression of cancer, invasion, and metastasis (44). Cancer cells can reactivate EMT pathways within their genomes and become more aggressive. EMT is linked to increased cancer stemness, which in turn drives cancer spread, recurrence, and resistance to treatment (43,45,46). It has been proposed in preclinical research that tumor cell EMT levels influence immunosuppression, with tumors with higher EMT levels being resistant to immunotherapy (47,48). The prognosis and therapeutic response can both be possibly predicted by characterizing EMT. Prior studies have shown that transcriptional factors involved in EMT cause immunosuppressive cells to infiltrate the tumor, creating an immunosuppressive microenvironment. Consequently, EMT in tumor cells is promoted by the immunosuppressive cells. Cancer progression is promoted by the interaction between EMT and immunosuppression indicating that EMT may develop as an important biomarker in the perspective of immunotherapy (49-51).

In a recent study on pleural mesothelioma, a panel of four EMT genes (COL5A2, ITGA, SPARC, and ACTA2) were identified and their overexpression in epithelioid mesothelioma was associated with poor prognosis, demonstrating that these genes have the potential to function as independent prognostic markers in this subtype of mesothelioma. In addition, overexpression of these genes was associated with the immunosuppressive microenvironment driven by EMT. Therefore, it was suggested that these genes may serve as predictive biomarkers for immunotherapy selection (52).

EMT was studied in MPeM, during a phase 2 single-center basket trial for the assessment of atezolizumab/bevacizumab in MPeM. The investigators concluded that transcriptome mesenchymal differentiation is a predictor of poor outcomes in MPeM when treated with atezolizumab/bevacizumab combination including those with epithelioid histology. Higher EMT gene scores were associated with poorer progression-free survival (PFS) on both atezolizumab/ bevacizumab and prior platinum pemetrexed chemotherapy. This finding is in line with prior evidence of EMT gene-signature score prognostic value in pleural mesothelioma (53,54). Moreover, they correlated baseline EMT gene expression scores with therapeutic resistance and response. A low EMT gene score was found to be a predictor of response even among patients with epithelioid subtype.

Both pleural and peritoneal mesotheliomas have been associated with sporadic and germline BAP1 mutations. MPeM with BAP1 mutation have 7-fold improved long-term survival (55,56). In a comprehensive genomic, transcriptomic, and proteomic investigation of 19 patients with MPeM, BAP1 deletions were linked to a more inflammatory tumor microenvironment. PDL1 and other immune checkpoint molecules were found to be highly expressed in BAP1-altered MPeM. The suggestion that BAP1 deletions may be utilized as a marker of ICIs responsiveness stems from these findings (57).

Multiple ICIs were studied as a single agent for the treatment of mesothelioma in general, including MPeM. The efficacy and toxicity of avelumab, an anti-PDL-1 was evaluated in phase 1b open-label study (JAVELIN Solid Tumor) in patients with mostly pleural and peritoneal mesothelioma that progressed after platinum and pemetrexed treatment. It was a single-arm study with 53 patients enrolled. Of note, the exact number MPeM patients was not reported. The study reported a low overall response rate (ORR) of about 10% all of which were partial responses (PR). Although modest in terms of response rate, responses were durable with a median duration of response of 15 months. Notably, the ORR was significantly higher in the subset of patients that had what they defined as high PDL1 expression (>5%). The disease control rate (DCR) was 58%. Median PFS was 4.1 months, and the median OS was 10.7 months. Safety data were reassuring, with no treatment-related deaths, and a 9% rate of grade 3 or 4 treatment-related adverse events. No inference can be made regarding the efficacy of avelumab in the MPeM since treatment outcomes were not analyzed by disease location (65).

CONFIRM was a placebo-controlled, double-blind, randomized trial to evaluate the efficacy and safety of nivolumab, an anti-PD-1 antibody in pretreated pleural or peritoneal mesothelioma, who had prior platinum-based chemotherapy in the first line. They enrolled 332 patients out of which 16 (5%) patients had MPeM. They were randomly assigned (2:1) to receive nivolumab at a flat dose of 240 mg every 2 weeks or placebo until disease progression or a maximum of 12 months. Patients were stratified by epithelioid vs. non-epithelioid histology. 88% of both arms had epithelioid histology, and 34% of patients had a tumor proportion score (TPS) of at least 1% indicating expression of PDL1. Nivolumab was a 3rd line treatment for more than half the patients. With a median follow-up of 11.6 months, nivolumab showed superior PFS with a median of 3 months vs. 1.8 months in the placebo arm. Median OS was also superior with 10.2 months in the nivolumab group vs. 6.9 months in the placebo group. In the nivolumab group, 41% of patients had serious adverse events (AEs), whereas 44% of placebo group patients did. There were no treatment-related deaths in either group. It's not clear from this trial if nivolumab performs better than single agent conventional chemotherapy in the 2nd line setting of treatment of MPeM given that it was compared to placebo. The majority of patients had pleural mesothelioma and a different trial that was conducted in pleural mesothelioma compared pembrolizumab to single agent chemotherapy and did not detect any difference in survival (66,67).

DETERMINE, was a negative phase 2b double-blind, placebo-controlled study that involved 105 centers across 19 countries of tremelimumab (antibody against CTLA-4) in patients with advanced malignant mesothelioma who had received prior one or two systemic treatments for advanced disease. Single-agent tremelimumab showed no survival benefit over placebo. Due to the fact that there were only 26 (5%) patients with MPeM included (out of a total of 571 patients with mesothelioma), subgroup analysis for efficacy specifically in MPeM was not feasible. It's possible that the use of a single anti-CTLA-4 agent in addition to the selection of a particular anti-CTLA-4 agent contributed to the lack of efficacy that was seen in the study (68).

Pembrolizumab was evaluated in a phase II study with 64 patients who had previously been treated for mesothelioma with no more than two prior lines of chemotherapy, including pemetrexed and platinum. Of those 64 patients, 8 patients (12.5%) had MPeM. Epithelioid was the predominant histology (76.6%), followed by biphasic (15.6%) and sarcomatoid (7.8%). In this study, the patients who had pleural mesothelioma demonstrated higher ORR than those who with MPeM (pleural 20%, peritoneal 12.5%). The response rate was analyzed by histology, with the highest response seen in sarcomatoid histology (40%). The response rate was reported to be 16% in epithelioid, and 10% in biphasic. It was noted that PDL1 expression was more common in the peritoneal subset with 25% of MPeM patients classified as having PDL1 high tumor (defined as TPS greater than or equal to 50%), 50% had PDL1 low (TPS 1 to 49%), and 25% of MPeM patients were PDL1 negative (TPS less than 1%). No threshold for PDL1 efficacy was determined from this study and there was no correlation found between the expression of PDL1 and response rate as a continuous metric. However, there was a trend to higher response rate and more durable PFS with increasing PDL1 expression (69).

A retrospective case series of 13 patients with MPeM who received pembrolizumab reported the treatment outcomes. All patients had received prior chemotherapy. The majority of the patient's histology was epithelioid (70%) followed by biphasic (15%), sarcomatoid, and desmoplastic (7.7% for each). Independent of the level of PDL1 expression or histology, pembrolizumab demonstrated an ORR of 18% all of which are PR. DCR was reported to be 81%, with a median PFS of 5.7 months. Median OS was reported to be 20.9 months. Of note, three patients had a PFS of more than 2 years (70).

In summary, due to the small number of MPeM patients treated in clinical trials, no definitive conclusion can be drawn about the efficacy of single agent immunotherapy in previously treated peritoneal mesothelioma. Response rates seem to be low and occur mainly in non-epithelioid histology. Additionally, it is unknown if PDL1 expression is predictive of benefit. Nevertheless, pembrolizumab remains a reasonable option on the rare occasion of high TMB and MSI-H MPeM.

Despite the drawbacks of extrapolating from pleural disease, combination immunotherapy has been used in MPeM due to its FDA approval and established efficacy in pleural mesothelioma. In malignant pleural mesothelioma, combination ICI-based treatment has shown encouraging success rates. In checkmate 743 which did not include any patients with MPeM, combination PD-1 and CTLA-4 inhibition with nivolumab and ipilimumab in treatment naïve malignant pleural mesothelioma significantly increased OS (median 18.1 months) compared to chemotherapy (median 14.1 months) (71).

The combination of tremelimumab plus durvalumab (anti-PDL1 monoclonal antibody) appears to result in clinically meaningful outcomes in mesothelioma. The single-arm phase II NIBIT-MESO-1 study that enrolled both treatment naïve and pretreated patients demonstrated an ORR of 28%, a DCR of 65%, and a median OS of 16.5 months. Despite the limited number of MPeM patients (2/40 (5%) enrolled), these results could support the use of tremelimumab combined with durvalumab in patients with MPeM given the OS benefit observed. The 4 years follow-up study suggested that retreatment may be safe and could still result in clinically meaningful outcomes (72,73).

A retrospective study conducted at a single institution in 29 patients with MPeM, reported an ORR of 19.2% following treatment with either a single agent ICI in 9 patients or a combination of nivolumab and ipilimumab in the remaining 20 patients. With a median follow-up of 9.8 months, 24 patients had discontinued ICIs (20 experienced progressive disease, 3 died, and 1 experienced toxic effects). The DCR was 65.4%. They reported a median PFS of 5.5 months and median OS of 19.1 months (74).

To recapitulate, no firm recommendation can be made about the use of combination ICIs in patients with MPeM at this time, given the absence of conclusive evidence that proves its safety and efficacy.

Chemoimmunotherapy showed favorable outcomes in pleural mesothelioma in the frontline setting. Patients with previously untreated, unresectable pleural mesothelioma were enrolled in PrE0505, a phase 2, single-arm, multicenter study. Patients received durvalumab (at a fixed dose of 1,120 mg intravenously) given once every 3 weeks in combination with pemetrexed and platinum at their standard doses for up to six cycles. Patients with stable or responding tumors after concurrent therapy continued on maintenance durvalumab for a maximum duration of 1 year. The primary endpoint of the study was OS which was met with combination of durvalumab/platinum/ pemetrexed also with a median OS of 21.1 months (75).

In MPeM, data are absent. In an intriguing case report, two patients with metastatic MPeM unresponsive to pemetrexed and platinum therapy, were treated with pembrolizumab in addition to pemetrexed /platinum, had a near-complete response in one patient and a remarkable partial response lasting 14 months in the other. Interestingly, both patients' tumors were PDL-1 negative and had low TMB. This provokes questions about the efficacy of chemoimmunotherapy in MPeM (76).

Vascular Endothelial Growth Factor (VEGF)/Vascular Endothelial Growth Factor Receptor (VEGFR) Pathway and Immune Checkpoint Inhibitor Combinations.

VEGF/VEGFR pathway inhibition in combination with chemotherapy resulted in minimal to no benefit at all in malignant pleural mesothelioma (64,77).

A cohort of 20 patients with MPeM was studied in an open-label, single-center basket trial for the assessment of atezolizumab/bevacizumab in a range of advanced uncommon malignancies. The study's objective was to evaluate the efficacy and safety of atezolizumab/bevacizumab in advanced MPeM who had previously failed systemic treatment with platinum-pemetrexed chemotherapy. There were only two patients with biphasic histology (10%), and all remaining patients had epithelioid histology. 12 patients (60%) had prior CRS and HIPEC in addition to systemic chemotherapy that all patients have received prior (53). An ORR of 40% was reported, with a median duration of response of 12.8 months. The 1-year OS rate of 86% seems meaningful when compared to the 1-year OS rate of (45-56%) reported in patients with prior platinum pemetrexed therapy (59,60,78). When compared to the previously reported 1-year OS rate of (45-56%) in patients with prior platinum pemetrexed treatment, the observed 1-year OS rate of 86% is considered significant.

Again, no correlation was found between atezolizumab/bevacizumab efficacy and PDL1 or TMB, and all tumors were microsatellite stable (MSS) which is consistent with the extreme rarity of MSI in mesothelioma (28). Since there was a role for EMT gene signature in predicting the prognosis of malignant pleural mesothelioma and diminishing ICIs responses in lung cancer, the EMT gene-signature scores were tested in this study. They demonstrated that transcriptome mesenchymal differentiation could predict poor outcomes with atezolizumab/bevacizumab therapy in MPeM, with higher EMT gene scores in mesenchymal phenotype being associated with resistance to atezolizumab/bevacizumab (54,79). Grade 3 AEs were reported in 50% of patients with the most common being hypertension (40%) and anemia (10%). Grade 3 immune-related AEs were reported in 10% of patients and required treatment discontinuation.

More definite evidence of efficacy and safety given the fact that there is a higher risk of intestinal perforation with bevacizumab in peritoneal mesothelioma patients is needed.