Abstract
Background
Programmed cell death protein 1 (PD-1) blockade improved the survival of patients with mismatch repair deficient (dMMR) and/or microsatellite instability-high (MSI-H) tumors, leading to tumor-agnostic approval of pembrolizumab in this population. Whether anti-programmed death (ligand)-1 [PD-(L)1] agents may achieve similar efficacy in dMMR/MSI-H metastatic gastric cancer (mGC) compared to metastatic colorectal cancer (mCRC) is unclear.
Materials and methods
We conducted a multicenter cohort study to collect data on patients with dMMR/MSI-H mGC or mCRC treated with anti-PD-(L)1 monotherapy globally at 17 tertiary cancer centers. Clinical features were balanced according to tumor type through the inverse probability of treatment weighting (IPTW) method. The primary endpoint was overall survival (OS), as evaluated from the first anti-PD-(L)1 administration.
Results
The cohort included 398 mCRC and 121 mGC patients, with a median follow-up of 34.6 and 25.1 months, respectively. The two populations differed for several baseline clinical features: patients with mCRC had younger age (60 versus 68 years, P < 0.001), better performance status (PS 0: 46% versus 34%, P = 0.062), higher frequency of primary tumor resection (82% versus 49%, P < 0.001) and liver metastases (38% versus 24%, P = 0.005), yet lower rates of distant nodal metastases (57% versus 83%, P < 0.001) and synchronous presentation (51% versus 76%, P < 0.001). After IPTW adjustment, patients with mGC showed no significant difference in progression-free survival (PFS) and OS compared to those with mCRC [PFS: hazard ratio (HR) 0.55, P = 0.077; OS: HR 0.65, P = 0.200].
Conclusions
Despite patients with dMMR/MSI-H mGC being enriched with poor prognostic factors as compared to the mCRC counterpart, anti-PD-(L)1 monotherapy’s efficacy appears similar in the two tumor types.
Key words: mismatch repair, microsatellite instability, colorectal cancer, gastric cancer, immunotherapy, inverse probability of treatment weighting
Highlights
-
•
We assessed PD-1 blockade’s efficacy in 519 patients with dMMR/MSI-H mGC and mCRC.
-
•
dMMR/MSI-H mGC patients were enriched with poor prognostic factors compared to mCRC patients.
-
•
Differences in baseline features were adjusted through IPTW.
-
•
PD-1 blockade showed similar efficacy in patients with dMMR/MSI-H mGC and mCRC.
Introduction
Tumors with mismatch repair deficiency (dMMR) or high microsatellite instability (MSI-H) present a unique mutational signature and harbor 10-100 times more mutations compared to the mismatch repair proficient or microsatellite stable counterpart.1,2 dMMR/MSI-H tumors can be found across several cancer types with different frequencies according to histotype and tumor stage.3,4 Although only 4%-5% of patients with metastatic colorectal cancer (mCRC) and metastatic gastric cancer (mGC) show dMMR/MSI-H, upfront testing is recommended because of the significant impact of MSI status on the therapeutic algorithms.5, 6, 7, 8 In fact, immune checkpoint inhibitors (ICIs) yielded durable responses in patients with dMMR/MSI-H tumors as shown by phase II trials with anti-programmed death (ligand)-1 [PD-(L)1] agents in multiple cancer types.9, 10, 11 These meaningful results led to the Food and Drug Administration ‘agnostic’ approval of pembrolizumab in patients with previously treated dMMR/MSI-H advanced solid tumors.12
However, the initial and updated results of KEYNOTE-158 highlighted that the peculiarities of each tumor type may significantly influence the efficacy of ICIs.9,10 The lack of sufficiently large cohorts of some tumor types and the disappointing results in specific tumor types such as MSI-H pancreatic cancer led to the restriction of the European Medicines Agency approval of pembrolizumab to dMMR/MSI-H mCRC, mGC, endometrial, small bowel and biliary tract cancers.13
Focusing on the most frequent gastrointestinal histotypes, programmed cell death protein 1 (PD-1) ± cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) blockade strategies are the standard of care in patients with dMMR/MSI-H mCRC based on the results of several studies including the KEYNOTE-177 phase III trial.14, 15, 16, 17, 18, 19 Given the relatively lower incidence of mGC compared to mCRC, no dedicated clinical trial with ICIs has been specifically conducted in patients with dMMR/MSI-H mGC. Nonetheless, several post hoc analyses of randomized controlled trials (RCTs) in mGC have shown the superior efficacy of anti-PD-(L)1-based regimens compared to chemotherapy in the dMMR/MSI-H subgroup and that MMR/MSI is the strongest predictive marker for anti-PD-(L)1-based therapy in this disease.20, 21, 22, 23, 24, 25, 26 These results are limited by their post hoc nature, the lack of stratification by MMR/MSI status in those RCTs and the small number of patients with dMMR/MSI-H status. Furthermore, the efficacy of ICIs in patients with dMMR/MSI-H mGC is yet largely unknown in the real-world setting.
Based on these considerations, we conducted a global retrospective study to compare the outcomes of two independent cohorts of patients with dMMR/MSI-H mGC versus dMMR/MSI-H mCRC treated with PD-1 blockade, adjusting for differences in baseline features with inverse probability of treatment weighting (IPTW).
Materials and methods
Patients’ population
This global retrospective study included patients with either dMMR/MSI-H mCRC or dMMR/MSI-H mGC treated with anti-PD-(L)1-based therapies from 17 institutions worldwide. Patients treated with an anti-CTLA-4 combination were excluded given the meaningful impact of dual ICI combination on outcomes and the negligible fraction of patients treated with this regimen in the mGC cohort.17, 18, 19 The following features were collected for each patient: sex, age, Eastern Cooperative Oncology Group performance status (ECOG PS), MMR and MSI status, primary tumor site, histotype, stage at diagnosis, primary tumor resection, synchronous presentation, number of metastatic sites, presence of liver, bone, lung and distant lymph node metastases, presence of ascites, previous systemic treatment for either local or advanced stage, type of PD-1 blockade regimen, PD-1 blockade-related adverse events (AEs). In patients with mGC, we also collected data on the PD-L1 combined positive score (CPS).
The dMMR/MSI-H status was assessed by local testing with immunohistochemistry and/or multiplex PCR according to international guidelines.
The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the Fondazione IRCCS Istituto Nazionale dei Tumori of Milan (INT 117/15).
Statistical analyses
Standard descriptive statistics were used to summarize clinical and biological patients’ characteristics according to the tumor entity. Differences in baseline features between the two groups were estimated using standardized differences (d value) to evaluate the imbalance between groups regardless of their size. Differences were considered negligible when d values were <0.1, small when d values were between 0.1 and 0.3, moderate when d values were between 0.3 and 0.5 and large when d values were >0.5.
The primary objective of the study was to compare overall survival (OS) between patients with mGC and mCRC in the IPTW-adjusted population.
OS was defined as the time from the first dose of ICI administration to death from any cause or last follow-up for alive patients. Progression-free survival (PFS) was defined as the time from the first dose of ICI administration to progression or death from any cause, whichever occurred first, and was evaluated as a secondary endpoint. Median follow-up time and median treatment duration time were calculated with the reverse Kaplan–Meier method.
A propensity score was calculated, defining the likelihood of being affected by mGC conditional to covariates. The propensity score calculation included all clinical and tumor variables available at the start of the treatment to avoid a potential imbalance of other parameters not correlated with the probability of being affected by mGC but with an unknown effect on the outcome. The features included in the propensity score calculation were the following: sex, age, ECOG PS, primary tumor resection, synchronous presentation, number of metastatic sites, liver, bone, lung and distant lymph node metastases and peritoneum involvement, previous systemic treatment for metastatic disease and number of previous systemic treatment lines. The obtained propensity score was then used to generate stabilized inverse probability of treatment weights, which were used to weigh each clinical feature, and thus the measured outcomes, of each patient in both groups. After weighting baseline characteristics, d values were re-calculated, and considered negligible if d values returned d < 0.1.
Differences between outcomes (OS and PFS) of mCRC and mGC patients were analyzed in the unadjusted and IPTW-adjusted populations. IPTW-adjusted Kaplan–Meier curves were calculated to graphically compare survivals among the groups.
The impact of tumor type together with grouping variables of interest was explored through standard or IPTW-adjusted log-rank and Cox regression models. Also, the impact of baseline clinical covariates on OS and PFS was assessed independently in each unadjusted population (mCRC and mGC), both in univariable and multivariable Cox regression models. Variables were included in multivariable models if significantly (P ≤ 0.05) associated with the outcome of interest in the univariate analysis. Results were expressed as hazard ratios (HRs) with the respective 95% confidence interval (CI).
Results
Patients’ characteristics
Supplementary Figure S1, available at https://doi.org/10.1016/j.esmogo.2023.100037, shows the study workflow. The two final cohorts included 398 patients with dMMR/MSI-H mCRC and 121 patients with dMMR/MSI-H mGC.
Several patient- and tumor-related characteristics were unbalanced between the two groups (Table 1). Firstly, patients with mGC had significantly older age (68 versus 60 years, P < 0.001) and worse PS (PS 1: 57.0% versus 46.2%, PS ≥2: 9.1% versus 7.8%, P = 0.062) as compared to those with mCRC. Furthermore, they had a lower prevalence of primary tumor resection (48.8% versus 81.7%, P < 0.001) and a higher frequency of synchronous presentation (76.0% versus 51.3%, P < 0.001). On the contrary, mGC patients had a lower frequency of liver metastases (24.0% versus 37.7%, P = 0.005) and a higher frequency of distant lymph node metastases (82.6% versus 57.0%, P < 0.001).
Table 1.
Patients and disease characteristics overall and according to tumor site of origin
| Characteristics | Total n = 519 | Colorectal n = 398 | Gastric n = 121 | Pa | db |
|---|---|---|---|---|---|
| Age | 62 (50-71) | 60 (48-70) | 68 (61-74) | <0.001 | 0.644 |
| Sex | 0.389 | 0.045 | |||
| Female | 258 (49.7) | 202 (50.7) | 56 (46.3) | ||
| Male | 261 (51.3) | 196 (49.3) | 65 (53.7) | ||
| ECOG PS | 0.062 | ||||
| 0 | 224 (43.2) | 183 (46.0) | 41 (33.9) | 0.121 | |
| 1 | 253 (48.7) | 184 (46.2) | 69 (57.0) | 0.108 | |
| ≥2 | 42 (8.1) | 31 (7.8) | 11 (9.1) | 0.013 | |
| Primary tumor resection | <0.001 | 0.329 | |||
| No | 135 (26.0) | 73 (18.3) | 62 (51.2) | ||
| Yes | 384 (74.0) | 325 (81.7) | 59 (48.8) | ||
| Synchronous metastases | <0.001 | 0.248 | |||
| No | 223 (43.0) | 194 (48.7) | 29 (24.0) | ||
| Yes | 296 (57.0) | 204 (51.3) | 92 (76.0) | ||
| Number of metastatic sites | 0.996 | 0.000 | |||
| 1 | 236 (45.5) | 181 (45.5) | 55 (45.5) | ||
| ≥2 | 283 (54.5) | 217 (54.5) | 66 (54.5) | ||
| Liver metastases | 0.005 | 0.137 | |||
| No | 340 (65.5) | 248 (62.3) | 92 (76.0) | ||
| Yes | 179 (34.5) | 150 (37.7) | 29 (24.0) | ||
| Bone metastases | 0.375 | 0.020 | |||
| No | 494 (95.2) | 377 (94.7) | 117 (96.7) | ||
| Yes | 25 (4.8) | 21 (5.3) | 4 (3.3) | ||
| Lung metastases | 0.380 | 0.035 | |||
| No | 428 (82.5) | 325 (81.7) | 103 (85.1) | ||
| Yes | 91 (17.5) | 73 (18.3) | 18 (14.9) | ||
| Lymph node metastases | <0.001 | 0.256 | |||
| No | 192 (37.0) | 171 (43.0) | 21 (17.4) | ||
| Yes | 327 (63.0) | 227 (57.0) | 100 (82.6) | ||
| Peritoneum involvement | 0.195 | ||||
| No | 329 (63.4) | 250 (62.8) | 79 (65.2) | 0.025 | |
| Yes, without ascites | 146 (28.1) | 118 (29.6) | 28 (23.1) | 0.065 | |
| Yes, with ascites | 44 (8.5) | 30 (7.5) | 14 (11.6) | 0.040 | |
| Prior systemic treatment | 0.812 | 0.010 | |||
| No | 116 (22.4) | 88 (22.1) | 28 (23.1) | ||
| Yes | 403 (77.7) | 310 (77.9) | 93 (76.9) | ||
| Treatment regimen | <0.001 | 0.830 | |||
| a-PD-(L)1 monotherapy | 509 (98.1) | 398 (100) | 111 (91.7) | ||
| a-PD-(L)1 + chemotherapy | 10 (2.9) | 0 (0) | 10 (8.3) | ||
| PD-L1 CPS ≥5 | c | c | |||
| <5 | 10 (15.2) | — | 10 (15.2) | ||
| ≥5 | 56 (84.8) | — | 56 (84.8) | ||
| NA | 453 | 398 | 55 (45.4) |
P values marked with bold indicate statistically significant P-values (≤0.05).
d values marked with bold indicate statistically significant differences in baseline features between the two groups (>0.1).
a-PD(L)-1, anti-programmed death (ligand)-1; CPS, combined positive score; ECOG PS, Eastern Cooperative Oncology Group performance status; NA, not available; OS, overall survival; PD-L1, programmed death-ligand 1; PFS, progression-free survival.
Chi-square/Fisher’s exact test or Wilcoxon rank test for categorical and continuous variables, respectively.
Standardized differences.
Not applicable due to the lack of data in the colorectal population.
Notably, as shown in Figure 1, age at ICI start (d value = 0.644) showed a large difference between the groups, whereas primary tumor resection (d value = 0.329) had a moderate difference and ECOG PS (PS 0: d value = 0.121, PS 1: d value = 0.108, PS 2: d value = 0.013), synchronous presentation (d value = 0.248), presence of liver metastases (d value = 0.137) and presence of nodal metastases (d value = 0.256) had small differences.
Figure 1.
Covariate balance plot before and after IPTW adjustment. The blue connected triangles and yellow connected dots represent the covariates’ absolute mean differences (d value) before and after IPTW adjustments; vertical dotted lines highlight cut-offs for negligible (<0.1), small (0.1-0.3), moderate (0.3-0.5) and large (>0.5) d values. ECOG PS, Eastern Cooperative Oncology Group performance status; IPTW, inverse probability of treatment weighting. aAbsolute standardized mean difference for continuous variables.
Primary outcomes
The median follow-up was 34.6 months [interquartile range (IQR) 21.0-53.9 months] in mCRC patients and 25.1 months (IQR 15.9-39.0 months) in mGC patients, with a median treatment duration of 14.3 months (IQR 2.9-24.8 months) and 11.5 months (IQR 4.2-20.4 months), respectively. At the data cut-off date (1 June 2023), 89 (22.4%) patients with mCRC and 33 (27.3%) patients with mGC were still receiving ICIs. The reasons for treatment discontinuation were AEs in 20 (5.0%) and 12 (9.9%) patients with mCRC and mGC, respectively, cancer progression in 161 (40.5%) and 39 (32.2%), cancer-related death (without radiological assessment of disease progression) in 11 (2.7%) and 1 (0.8%), surgery, in locally advanced or oligometastatic tumors deemed resectable after treatment response, in 14 (3.5%) and 4 (3.3%), treatment completion, defined as a complete or a sustained partial response after 2 years of treatment, in 49 (12.3%) and 20 (16.5%) and other causes in 7 (1.8%) and 12 (9.9%).
In the unadjusted population, patients with dMMR/MSI-H mGC exhibited no significant difference in terms of OS (2-year OS: 63.7% versus 66.7%; HR 1.14, 95% CI 0.89-1.60, P = 0.434) and PFS (2-year PFS: 56.0% versus 51.1%; HR 0.86, 95% CI 0.64-1.17, P = 0.338) compared to patients with dMMR/MSI-H mCRC (Figure 2).
Figure 2.
Kaplan-Meier curves for progression free survival and overall survival before IPTW adjustment. Kaplan–Meier curves for progression-free survival (A) and overall survival (B) in the cohort of patients with dMMR/MSI-H mGC or MMR/MSI-H before adjusting for the difference in baseline features. CI, confidence interval; dMMR, mismatch repair deficient; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; mCRC, metastatic colorectal cancer; mGC, metastatic gastric cancer; MSI-H, microsatellite instability-high; NA, not available; OS, overall survival; PFS, progression-free survival; Ref, reference.
Supplementary Tables S1 and S2, available at https://doi.org/10.1016/j.esmogo.2023.100037, show uni- and multivariable Cox models for both PFS and OS for patients with dMMR/MSI-H mGC or dMMR/MSI-H mCRC, respectively. The ECOG PS had an independent effect on OS and PFS in both groups, while bone metastases had an independent effect on OS in both groups. In the mCRC group, lung metastases had an independent effect on PFS and OS, while the presence of ascites had an independent effect on OS. In the mGC group, primary tumor resection and the presence of multiple metastatic sites also had an independent association with worse OS and PFS, while the combined treatment regimen of anti-PD-(L)1 and chemotherapy was associated with poorer PFS as compared to immunotherapy alone. Of note, we did not highlight an independent association between CPS and survival outcomes at multivariable analysis, though with a trend toward greater benefit in cases with CPS ≥5. Lastly, our study evidenced no significant association between previous systemic treatment for metastatic disease and survival outcomes in either mCRC or mGC patients.
Inverse probability of treatment weighting adjustment
After IPTW adjustment, some features still retained small-to-moderate differences. As shown in Figure 1, only age of ICI start (d value = 0.401) showed moderate differences between the IPTW-adjusted groups, while ECOG PS (PS 0: d value = 0.126, PS 1: d value = 0.093, PS 2: d value = 0.033), synchronous presentation (d value = 0.158) and presence of liver metastases (d value = 0.145) showed small differences.
In the IPTW-adjusted population, the two groups did not show any statistically significant difference in terms of OS and PFS. Notably, patients with dMMR/MSI-H mGC had an improved OS (estimated 2-year OS: 79.1% versus 67.7%; HR 0.65, 95% CI 0.33-1.26, P = 0.2) and PFS (estimated 2-year PFS: 67.3% versus 51.1%; HR 0.55, 95% CI 0.28-1.07, P = 0.077) compared to mCRC patients, although this trend did not reach the threshold of statistical significance (Figure 3). Supplementary Table S3, available at https://doi.org/10.1016/j.esmogo.2023.100037, shows the multivariable model for both PFS and OS including the features still unbalanced after the IPTW adjustment. While ECOG PS maintained an independent effect on both PFS and OS, still no significant difference emerged between mGC and mCRC.
Figure 3.
Kaplan-Meier curves for progression free survival and overall survival after IPTW adjustment. Kaplan–Meier curves for progression-free survival (A) and overall survival (B) in the cohort of patients with dMMR/MSI-H mGC or MMR/MSI-H adjusted for the difference in baseline features through the IPTW method. CI, confidence interval; dMMR, mismatch repair deficient; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; IPTW, inverse probability of treatment weighting; mCRC, metastatic colorectal cancer; mGC, metastatic gastric cancer; MSI-H, microsatellite instability-high; NA, not available; OS, overall survival; PFS, progression-free survival; Ref, reference.
Discussion
We sought to compare the efficacy of PD-1 blockade in a cohort of patients with dMMR/MSI-H mGC versus a cohort of patients with dMMR/MSI-H mCRC, by adjusting for the differences in baseline features, to determine whether MSI could predict the tumor response regardless of the tumor site of origin.
Our data confirmed that PD-1 blockade is an effective treatment strategy in patients with MSI-H/dMMR mGC, reporting a 2-year OS over 60%. These outcomes are consistent with previous subgroup analyses from published RCTs and are considerably better compared to the overall population of patients with mGC.21, 22, 23
In the unadjusted populations, patients with mGC did not present a statistically significant difference in OS and PFS compared to patients with mCRC, despite the large differences in baseline features.
After the IPTW adjustment, although not fully balanced, the two groups presented only small-to-intermediate differences in baseline features and no meaningful difference in OS and PFS. Notably, patients in the mGC group even presented non-significant improvements in OS and PFS compared to those in the mCRC group.
These results also underline how baseline features associated with dMMR/MSI-H mGC may have a strong influence on patients’ survival outcomes. Although poor clinical characteristics are obviously associated with a worse prognosis, some features such as ascites, bone metastases and ECOG PS could also contribute to the onset of immune-resistance mechanism and early progression.27,28 Also, the rapid health status decline may preclude exposure to potentially effective and/or sequential therapies. On this note, in our study, previous systemic treatment for metastatic disease, despite being usually associated with a worse prognosis, did not affect patients’ survival outcomes, reinforcing the suggestion that MSI/dMMR holds the strongest predictive value to ICI response. For these reasons, immediate access to the most effective treatments according to tumor and clinical features is crucial.
In our cohort, only a small proportion of patients (8%) in the dMMR/MSI-H mGC cohort received an anti-PD-1 agent combined with chemotherapy (Table 1); therefore, a robust assessment of the role of chemotherapy combined with ICIs in this population could not be conclusive. Post hoc analyses of RCTs did not show any meaningful difference between anti-PD-1 therapy alone or combined with chemotherapy in dMMR/MSI-H mGC patients.20,21 Of note, our analysis showed that the addition of chemotherapy was even associated with poorer PFS. If confirmed in larger studies, this would suggest that patients with dMMR/MSI-H mGC may risk overtreatment and exposure to unnecessary or even potentially detrimental chemotherapy. Indeed, international guidelines currently recommend first-line use of an anti-PD-1 agent in combination with chemotherapy in patients with mGC according to PD-L1 CPS, regardless of MSI-H status.7,8 On the other side, although patients with dMMR/MSI-H mGC mostly show high tumor PD-L1 expression, some of them may have CPS <5 and could miss the opportunity of receiving first-line ICI-based therapy. In this regard, our results, despite the small portion of patients with CPS <5, showed a non-significant association between CPS and survival outcomes in the dMMR/MSI mGC population in the multivariable models accounting for other clinical factors.
Patients who received dual PD-1/CTLA-4 blockade were excluded from our analysis, since only a negligible fraction of patients were treated with this combination in the mGC cohort. While combined anti-PD-1/anti-CTLA-4 regimens showed promising outcomes in the dMMR/MSI-H mCRC patients, the evidence of the ICI combination in the dMMR/MSI-H mGC patients is once again limited to post hoc analyses of small groups of patients.17, 18, 19,29 At the moment, the efficacy of nivolumab plus low-dose ipilimumab in untreated metastatic MSI-H/dMMR GC/gastroesophageal junction cancer is being investigated by the NO LIMIT phase II non-randomized trial (WJOG13320G/CA209-7W7), which may provide further evidence regarding the role of combined immunotherapy in this population.30
From a future perspective, biomarkers should be implemented to predict sensitivity or resistance to ICIs beyond the intrinsic life expectancy associated with specific clinical features of patients with dMMR/MSI-H mGC.27,31 For instance, tumor mutational burden, gene expression and mutational signatures, which are being explored in translational research settings, could be investigated in real-world datasets and in clinical trials.32, 33, 34, 35
Our study has several limitations. Firstly, due to its retrospective nature, our work cannot provide the same level of evidence of RCTs, even accounting for the IPTW analysis. Furthermore, although the IPTW-adjusted populations did not have any large difference in baseline features, they still presented some not negligible differences with a d value >0.1. The largely inferior number of patients in the mGC group also determined equally increased patients’ weights in the IPTW-adjusted population which could potentially result in overestimations of the survival outcomes in the mGC cohort. Moreover, although we included previous treatment as one of the baseline features, we did not analyze whether the number of previous lines could be associated with the survival outcomes. Another potential limitation is represented by the relatively short follow-up duration, especially considering that patients with dMMR/MSI-H cancer treated with ICIs can achieve sustained tumor responses and long-term survival. Finally, we acknowledged that the low rate of patients treated with the combination regimen (chemotherapy + PD-1 blockade) does not allow us to properly assess any potential differences in terms of survival. Similarly, considering the small portion of patients with CPS <5, we were not able to definitely determine the predictive or prognostic role of CPS in the dMMR/MSI-H mGC population.
Conclusion
Despite patients with dMMR/MSI-H mGC being enriched with poor prognostic factors compared to the mCRC counterpart, the efficacy of ICIs appears similar in the two tumor types. PD-1 blockade is an effective strategy in dMMR/MSI-H mGC and patients should be given the opportunity to receive anti-PD-(L)1 agents (with or without chemotherapy) as soon as possible during the disease course and before the deterioration of health conditions. RCTs should be specifically conducted in patients with dMMR/MSI-H mGC and the relevant prognostic variables investigated here should be considered as key clinical subgroups.
Acknowledgments
Funding
This work has been supported by Fondazione AIRC per la Ricerca sul Cancro ETS (Project ID: IG 2019 23624 - R19004 PI: Md. Filippo Pietrantonio).
Disclosure
KS reports receiving personal fees for consulting and advisory roles from Bristol Myers Squibb, Takeda, Ono Pharmaceutical, Novartis, Daiichi Sankyo, Amgen, Boehringer Ingelheim, Merck Pharmaceutical, Astellas, Guardant Health Japan, Janssen, AstraZeneca, Zymeworks Biopharmaceuticals, ALX Oncology Inc. and Bayer; receiving honoraria from Bristol-Myers Squibb, Ono Pharmaceutical, Janssen, Eli Lilly, Astellas and AstraZeneca; and receiving research funding (all to institution) from Astellas, Ono Pharmaceutical, Daiichi Sankyo, Taiho Pharmaceutical, Chugai, Merck Pharmaceutical, Amgen, Eisai, PRA Health Sciences and Syneos Health, outside the submitted work. RCo reports receiving personal fees for consulting and advisory roles from Exeliom Biosciences, Enterome and Bristol Myers Squibb; receiving honoraria from Pierre Fabre, Bristol Myers Squibb, Servier, Mylan and AstraZeneca/Daiichi Sankyo; and receiving funding for travel, accommodations and expenses from Amgen, Bristol Myers Squibb, Mylan and Servier. SL reports participation in advisory board for Amgen, Astellas, Astra Zeneca, Bayer, Bristol-Myers Squibb, Daiichi-Sankyo, GSK, Incyte, Lilly, Merck Serono, MSD, Servier and Takeda; receiving personal honoraria as invited speaker from Amgen, Bristol-Myers Squibb, Incyte, GSK, Lilly, Merck Serono, MSD, Pierre-Fabre, Roche and Servier; and receiving research funding (all to institution) from Amgen, Astellas, Astra Zeneca, Bayer, Bristol-Myers Squibb, Daiichi-Sankyo, GSK, Incyte, Lilly, Merck Serono, MSD, Servier and Takeda outside the submitted work. CC reports participation in advisory board from MSD, Nordic Pharma, Pierre Fabre, Roche and Takeda; receiving personal honoraria as invited speaker from Amgen, Bayer, Merck Serono and Servier; and receiving research funding (all to institution) from Bayer, Hutchinson, Merck’s, Roche, Seagen and Servier outside the submitted work. MEE reports receiving personal honoraria for advisory role, travel grants and research grants from Amgen, Bayer, Hoffman-La Roche, Merck Serono, MSD, Novartis, Organon, Pfizer, Pierre Fabre, Sanofi, Seagen International GmbH, Servier and Takeda; receiving research funding (all to institution) from Amgen Inc, Array Biopharma Inc, AstraZeneca Pharmaceuticals LP, BeiGene, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Debiopharm International SA, Genentech Inc, HalioDX SAS, Hoffmann-La Roche Ltd, Hutchison MediPharma International, Janssen-Cilag SA, MedImmune, Menarini, Merck Health KGAA, Merck Sharp & Dohme, Merus NV, Mirati, Novartis Farmacéutica SA, Pfizer, Pharma Mar, Sanofi Aventis Recherche & Développement, Servier and Taiho Pharma USA Inc outside the submitted work; and participating as a volunteer member of the ASCO Annual Meeting Scientific Program Committee: Developmental Therapeutics Immunotherapy Track Leader for the 2023-2024 term; participating as speaker of the ESMO Academy and ESMO Congress 2025 Scientific Steering Committee Track Chair; and being assigned as coordinator of the Sociedad Española de Oncología Médica (SEOM) + MIR Section of Residents and Young Assistants and member of the Scientific Committee of the next SEOM2024 and SEOM2025 Congresses. JC reports receiving personal fees for consulting and advisory roles from Bristol-Myers Squibb, Astellas, Roche, Guardant Health, Yiviva, Daiichi-Sankyo and Merck; receiving honoraria as invited speaker from Bristol-Myers Squibb, Astellas, Roche, Guardant Health Merck; receiving employment and employee stock options from Amgen; and receiving research funding (all to institution) from Merck and Brooklyn Immunotherapeutics outside the submitted work. MF reports receiving personal fees for consulting and advisory roles from AbbVie, AstraZeneca, Bayer Corporation, Bristol Myers Squibb, Eisai, Entos, Janssen, Merck, Mirati Therapeutics, Nouscom, Pfizer, Roche/Genentech and Taiho Oncology; and receiving research funding (all to institution) from AgenusBio, Bristol Myers Squibb, Genentech/imCor and Verastem outside the submitted work. SJK reports receiving personal fees and non-financial support from Astellas, Merck, BMS, Mersana, Sanofi Aventis, AstraZeneca, Daiichi Sankyo, Coherus, Exact Sciences, Servier, Eli Lilly and Amgen and other support from Turning Point Therapeutics and Nuvalent outside the submitted work. SM reports receiving honoraria as invited speaker from Basilea, Bicara, Daiichi-Sankyo, Elevation Oncology, Purple Oncology, Amgen, Novartis and Natera; receiving research funding (all to institution) from Conquer Cancer Foundation; and receiving research travel support from AstraZeneca outside the submitted work. LF reports participation in advisory board for MSD, AstraZeneca, Incyte, Taiho, Servier, Daiichi Sankyo, Lilly and Astellas; receiving personal honoraria as invited speaker from Incyte, Bristol Myers Squibb and Lilly; and receiving research funding (all to institution) from MSD, Bristol Myers Squibb, AstraZeneca, Incyte, BeiGene, Astellas, Daiichi Sankyo and Roche outside the submitted work. LS reports receiving personal fees as consulting role for Pierre-Fabre, AstraZeneca, Bayer, Servier, Merck and Amgen. LS reports receiving personal honoraria as invited speaker and consultant’s fee from MSD, Astra-Zeneca, Servier, Bayer, Merck, Amgen, Pierre-Fabre, Takeda, GSK, Incyte, Bristol Myers Squibb and Lilly. FB reports receiving personal fees for consulting and advisory roles from Bristol Myers Squibb, AAA, Lilly, Servier and Eisai. MJO reports receiving personal fees for consulting and advisory roles from 3D Medicines, Array BioPharma, Bristol-Myers Squibb, Eisai, Gritstone Bio, Janssen, MedImmune, Merck, Novartis, Pfizer, Promega, Roche/Genentech and Spectrum Pharmaceuticals; and receiving research funding (all to institution) from Bristol-Myers Squibb, MedImmune, Merck and Roche outside the submitted work. TA reports receiving personal fees for consulting and advisory roles from Abbvie, Astellas, Aptitude Health, Bristol Myers Squibb, Gritstone Oncology, Gilead, GlaxoSmithKline, Merck & Co. Inc., Nordic Oncology, Seagen, Servier, Takeda and Transgène; receiving honoraria as invited speaker from Bristol Myers Squibb, GlaxoSmithKline, Merck & Co. Inc., Merck Serono, Pierre Fabre, Roche, Sanofi, Seagen, Servier and Takeda; and receiving support for meetings from Merck & Co. Inc. and Servier. FP reports receiving personal fees for consulting and advisory roles from Amgen, Merck-Serono, MSD, Bayer, Astellas, Takeda, Servier, GlaxoSmithKline, Pierre-Fabre, Ipsen and J&J; and receiving research funding (all to institution) from Lilly, BMS, AstraZeneca, Incyte, Amgen and Agenus outside the submitted work. All other authors have declared no conflicts of interest.
Given their role as Associate Editor and Editoiral Board member, Thierry André and Sara Lonardi had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Florian Lordick, Editor-in-Chief, of the Journal.
Data sharing
All data included in this publication will be shared by the authors upon specific personal request.
Supplementary data
References
- 1.Alexandrov L.B., Nik-Zainal S., Wedge D.C., et al. Australian Pancreatic Cancer Genome Initiative, ICGC Breast Cancer Consortium, ICGC MMML-Seq Consortium. Signatures of mutational processes in human cancer. Nature. 2013;500(7463):415–421. doi: 10.1038/nature12477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dudley J.C., Lin M.T., Le D.T., Eshleman J.R. Microsatellite instability as a biomarker for PD-1 blockade. Clin Cancer Res. 2016;22(4):813–820. doi: 10.1158/1078-0432.CCR-15-1678. [DOI] [PubMed] [Google Scholar]
- 3.Latham A., Srinivasan P., Kemel Y., et al. Microsatellite instability is associated with the presence of Lynch syndrome pan-cancer. J Clin Oncol. 2019;37(4):286–295. doi: 10.1200/JCO.18.00283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bonneville R., Krook M.A., Kautto E.A., et al. Landscape of microsatellite instability across 39 cancer types. JCO Precis Oncol. 2017;1:1–15. doi: 10.1200/PO.17.00073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Morris V.K., Kennedy E.B., Baxter N.N., et al. Treatment of metastatic colorectal cancer: ASCO guideline. J Clin Oncol. 2023;41(3):678–700. doi: 10.1200/JCO.22.01690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Van Cutsem E., Cervantes A., Nordlinger B., Arnold D., ESMO Guidelines Working Group Metastatic colorectal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2014;25:iii1–iii9. doi: 10.1093/annonc/mdu260. [DOI] [PubMed] [Google Scholar]
- 7.Shah M.A., Kennedy E.B., Alarcon-Rozas A.E., et al. Immunotherapy and targeted therapy for advanced gastroesophageal cancer: ASCO guideline. J Clin Oncol. 2023;41(7):1470–1491. doi: 10.1200/JCO.22.02331. [DOI] [PubMed] [Google Scholar]
- 8.Lordick F., Carneiro F., Cascinu S., et al. Gastric cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2022;33(10):1005–1020. doi: 10.1016/j.annonc.2022.07.004. [DOI] [PubMed] [Google Scholar]
- 9.Marabelle A., Le D.T., Ascierto P.A., et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer: results from the phase II KEYNOTE-158 study. J Clin Oncol. 2020;38(1):1–10. doi: 10.1200/JCO.19.02105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Maio M., Ascierto P.A., Manzyuk L., et al. Pembrolizumab in microsatellite instability high or mismatch repair deficient cancers: updated analysis from the phase II KEYNOTE-158 study. Ann Oncol. 2022;33(9):929–938. doi: 10.1016/j.annonc.2022.05.519. [DOI] [PubMed] [Google Scholar]
- 11.Le D.T., Uram J.N., Wang H., et al. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372(26):2509–2520. doi: 10.1056/NEJMoa1500596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.US F.D.A. FDA grants accelerated approval to pembrolizumab for first tissue/site agnostic indication. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-grants-accelerated-approval-pembrolizumab-first-tissuesite-agnostic-indication Available at.
- 13.MERCK. European Commission Approves Merck’s KEYTRUDA® (pembrolizumab) for patients with microsatellite instability-high (MSI-H) or deficient mismatch repair (dMMR) tumors in five different types of cancer. https://www.merck.com/news/european-commission-approves-mercks-keytruda-pembrolizumab-for-patients-with-microsatellite-instability-high-msi-h-or-deficient-mismatch-repair-dmmr-tumors-in-five-different-types/ Available at.
- 14.André T., Shiu K.K., Kim T.W., et al. Pembrolizumab in microsatellite-instability–high advanced colorectal cancer. N Engl J Med. 2020;383(23):2207–2218. doi: 10.1056/NEJMoa2017699. [DOI] [PubMed] [Google Scholar]
- 15.Le D.T., Kim T.W., Van Cutsem E., et al. Phase II open-label study of pembrolizumab in treatment-refractory, microsatellite instability–high/mismatch repair-deficient metastatic colorectal cancer: KEYNOTE-164. J Clin Oncol. 2020;38(1):11–19. doi: 10.1200/JCO.19.02107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Overman M.J., McDermott R., Leach J.L., et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal cancer (CheckMate 142): an open-label, multicentre, phase 2 study. Lancet Oncol. 2017;18(9):1182–1191. doi: 10.1016/S1470-2045(17)30422-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.André T., Lonardi S., Wong K.Y.M., et al. Nivolumab plus low-dose ipilimumab in previously treated patients with microsatellite instability-high/mismatch repair-deficient metastatic colorectal cancer: 4-year follow-up from CheckMate 142. Ann Oncol. 2022;33(10):1052–1060. doi: 10.1016/j.annonc.2022.06.008. [DOI] [PubMed] [Google Scholar]
- 18.Overman M.J., Lonardi S., Wong K.Y.M., et al. Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-high metastatic colorectal cancer. J Clin Oncol. 2018;36(8):773–779. doi: 10.1200/JCO.2017.76.9901. [DOI] [PubMed] [Google Scholar]
- 19.Lenz H.J., Van Cutsem E., Luisa Limon M., et al. First-line nivolumab plus low-dose ipilimumab for microsatellite instability-high/mismatch repair-deficient metastatic colorectal cancer: the phase II CheckMate 142 study. J Clin Oncol. 2022;40(2):161–170. doi: 10.1200/JCO.21.01015. [DOI] [PubMed] [Google Scholar]
- 20.Pietrantonio F., Randon G., Di Bartolomeo M., et al. Predictive role of microsatellite instability for PD-1 blockade in patients with advanced gastric cancer: a meta-analysis of randomized clinical trials. ESMO Open. 2021;6(1) doi: 10.1016/j.esmoop.2020.100036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shitara K., Van Cutsem E., Bang Y.J., et al. Efficacy and safety of pembrolizumab or pembrolizumab plus chemotherapy vs chemotherapy alone for patients with first-line, advanced gastric cancer: the KEYNOTE-062 phase 3 randomized clinical trial. JAMA Oncol. 2020;6(10):1571. doi: 10.1001/jamaoncol.2020.3370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shitara K., Özgüroğlu M., Bang Y.J., et al. Pembrolizumab versus paclitaxel for previously treated, advanced gastric or gastro-oesophageal junction cancer (KEYNOTE-061): a randomised, open-label, controlled, phase 3 trial. Lancet. 2018;392(10142):123–133. doi: 10.1016/S0140-6736(18)31257-1. [DOI] [PubMed] [Google Scholar]
- 23.Moehler M., Shitara K., Garrido M., et al. LBA6_PR Nivolumab (nivo) plus chemotherapy (chemo) versus chemo as first-line (1L) treatment for advanced gastric cancer/gastroesophageal junction cancer (GC/GEJC)/esophageal adenocarcinoma (EAC): first results of the CheckMate 649 study. Ann Oncol. 2020;31:S1191. [Google Scholar]
- 24.Janjigian Y.Y., Shitara K., Moehler M., et al. First-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trial. Lancet. 2021;398(10294):27–40. doi: 10.1016/S0140-6736(21)00797-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Janjigian Y.Y., Shitara K., Moehler M.H., et al. Nivolumab (NIVO) plus chemotherapy (chemo) vs chemo as first-line (1L) treatment for advanced gastric cancer/gastroesophageal junction cancer/esophageal adenocarcinoma (GC/GEJC/EAC): 3-year follow-up from CheckMate 649. J Clin Oncol. 2023;41(suppl 4):291. [Google Scholar]
- 26.Moehler M., Dvorkin M., Boku N., et al. Phase III trial of avelumab maintenance after first-line induction chemotherapy versus continuation of chemotherapy in patients with gastric cancers: results from JAVELIN Gastric 100. J Clin Oncol. 2021;39(9):966–977. doi: 10.1200/JCO.20.00892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mazzoli G., Cohen R., Lonardi S., et al. Prognostic impact of performance status on the outcomes of immune checkpoint inhibition strategies in patients with dMMR/MSI-H metastatic colorectal cancer. Eur J Cancer. 2022;172:171–181. doi: 10.1016/j.ejca.2022.05.044. [DOI] [PubMed] [Google Scholar]
- 28.Fucà G., Cohen R., Lonardi S., et al. Ascites and resistance to immune checkpoint inhibition in dMMR/MSI-H metastatic colorectal and gastric cancers. J Immunother Cancer. 2022;10(2) doi: 10.1136/jitc-2021-004001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Shitara K., Ajani J.A., Moehler M., et al. Nivolumab plus chemotherapy or ipilimumab in gastro-oesophageal cancer. Nature. 2022;603(7903):942–948. doi: 10.1038/s41586-022-04508-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kawakami H., Hironaka S., Esaki T., et al. An investigator-initiated phase 2 study of nivolumab plus low-dose ipilimumab as first-line therapy for microsatellite instability-high advanced gastric or esophagogastric junction cancer (NO LIMIT, WJOG13320G/CA209-7W7) Cancers (Basel) 2021;13(4):805. doi: 10.3390/cancers13040805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Randon G., Aoki Y., Cohen R., et al. Outcomes and a prognostic classifier in patients with microsatellite instability-high metastatic gastric cancer receiving PD-1 blockade. J Immunother Cancer. 2023;11(6) doi: 10.1136/jitc-2023-007104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kwon M., An M., Klempner S.J., et al. Determinants of response and intrinsic resistance to PD-1 blockade in microsatellite instability-high gastric cancer. Cancer Discov. 2021;11(9):2168–2185. doi: 10.1158/2159-8290.CD-21-0219. [DOI] [PubMed] [Google Scholar]
- 33.Manca P., Corti F., Intini R., et al. Tumour mutational burden as a biomarker in patients with mismatch repair deficient/microsatellite instability-high metastatic colorectal cancer treated with immune checkpoint inhibitors. Eur J Cancer. 2023;187:15–24. doi: 10.1016/j.ejca.2023.03.029. [DOI] [PubMed] [Google Scholar]
- 34.Gallois C., Landi M., Taieb J., et al. Transcriptomic signatures of MSI-high metastatic colorectal cancer predict efficacy of immune checkpoint inhibitors. Clin Cancer Res. 2023;29(18):3771–3778. doi: 10.1158/1078-0432.CCR-22-3964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Westcott P.M.K., Muyas F., Hauck H., et al. Mismatch repair deficiency is not sufficient to elicit tumor immunogenicity. Nat Genet. 2023;55(10):1686–1695. doi: 10.1038/s41588-023-01499-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.