Whether there is an association between SWI/SNF genomic alterations in tumors and response to immune checkpoint inhibitors (ICI) remains unclear because prior studies have focused on either an individual gene or a predefined set of genes. Herein, using mutational and clinical data from 832 ICI-treated patients who underwent whole-exome sequencing, including sequencing of all 31 genes of the SWI/SNF complex, we found that SWI/SNF complex alterations were associated with significantly improved overall survival (OS) in melanoma, clear-cell renal cell carcinoma, and gastrointestinal cancer, as well as improved progression-free survival (PFS) in non–small cell lung cancer. Including tumor mutational burden as a variable, the multivariate Cox regression analysis showed SWI/SNF genomic alterations had prognostic value in melanoma [HR, 0.63 (95% confidence interval, CI, 0.47–0.85), P = 0.003], clear-cell renal cell carcinoma [HR, 0.62 (95% CI, 0.46–0.85), P = 0.003], and gastrointestinal cancer [HR, 0.42 (95% CI, 0.18–1.01), P = 0.053]. Furthermore, we used the random forest method for variable screening, identifying 14 genes as a SWI/SNF signature for potential clinical application. Significant correlations were observed between SWI/SNF signature alterations and improved OS and PFS in all cohorts. This suggests that SWI/SNF gene alterations are associated with better clinical outcomes in ICI-treated patients and may serve as a predictive marker for ICI therapy in multiple cancers.

In the last decade, cancer immunotherapy, especially immune checkpoint inhibitor (ICI) therapy, has become an indispensable part of comprehensive cancer therapy with unprecedented benefits in multiple cancers. ICIs can stimulate durable anticancer immune responses or enhance tumor cell killing by targeting immune checkpoint proteins, such as CTL-associated protein 4 (CTLA4), programmed cell death-1 (PD-1), and programmed death ligand 1 (PD-L1; ref. 1). However, only a subset of patients with cancer (∼20%–30%) experience prolonged clinical effects with ICI therapy, and the efficacy varies among different tumor types (2, 3). Given the high cost and risk of toxicity, the need to identify robust biomarkers that predict individual clinical responses or resistance to ICI therapy has become a clinical challenge.

Tumor mutational burden (TMB) is a predictive biomarker that the FDA has approved for selecting patients with unresectable or metastatic solid tumors for treatment with pembrolizumab (4). However, new research evaluating its performance has uncovered limitations. TMB assessment is not standardized across clinical studies and varies in genome coverage, sequencing platform, and bioinformatic pipeline. Meanwhile, targeted sequencing, the most used method in clinical practice, usually overestimates TMB (5). In addition, there have been concerns raised about the applicable cancer types, determination of the cutoff, and the ancestry-specific differences (6, 7). Moreover, owing to the internal heterogeneity of TMB-high tumors, other stratification indicators are needed to further screen the appropriate population that is most likely to respond. Therefore, other biomarkers predicting the response to ICI therapy need to be identified, thereby enabling personalized immunotherapy treatment.

Emerging preclinical and clinical research has focused on an association between SWItch/sucrose nonfermentable (SWI/SNF) complex genomic alterations and the efficacy of ICIs in multiple tumors (8, 9). However, it remains unclear whether SWI/SNF genomic alterations can predict response to ICI therapy. Preclinical studies have revealed that SWI/SNF inactivation is associated with increased mutation load, elevated numbers of tumor-infiltrating lymphocytes, and enhanced sensitivity to ICIs in some cancer types (10–13). However, some researchers reported contradictory findings that SWI/SNF inactivation leads to a nonimmunogenic tumor phenotype and poor tumor immunity (14, 15). Clinical research has associated SWI/SNF complex alterations with immunotherapy efficacy in urothelial carcinoma (13), pancreatic cancer (16), melanoma (17), and renal cell carcinoma (18, 19). Nevertheless, two other clinical trials did not observe an apparent association between PBRM1 mutations and improved progression-free survival (PFS) in patients with renal cell carcinoma treated with ICIs (20, 21). Similarly, partial retrospective studies with large sample sizes did not find a consistent association between SWI/SNF variants and improved clinical outcomes following ICI therapy (9, 22, 23), but some studies have found that patients with SWI/SNF-mutated tumors treated with an ICI have a statistically significant improvement in overall survival (OS; refs. 17, 24). Further research is required to determine whether SWI/SNF variants can be promising biomarkers for ICI therapy.

We noticed that these prior studies on the SWI/SNF complex focused on either an individual gene or a predefined set of SWI/SNF genes (9, 17, 24–26). However, the complex is composed of protein products of 31 genes. Therefore, previous conclusions may not provide the whole picture because they did not assess variations in all the coding genes. Data from multiple ICI-treated patient cohorts that had undergone whole-exome sequencing (WES) were collected to explore the association between SWI/SNF genomic alterations and responses to ICI therapy. We used the random forest method to select genes to generate a SWI/SNF signature for potential clinical application.

Patient cohorts

We collected mutational and clinical data from seven cohorts of patients treated with an ICI (Supplementary Table S1). These cohorts included 1,630 patients from the TMB MSKCC 2018 cohort whose tumors underwent targeted next-generation sequencing (MSK-IMPACT), including 138 patients with head and neck cancer, 317 patients with melanoma, 110 patients with colorectal cancer, and 43 patients with breast cancer (Supplementary Table S2; ref. 27). Together, the six other cohorts included 832 patients who underwent WES before ICI therapy, including 151 patients with melanoma from the Mixed Allen 2018 cohort (Supplementary Table S3; ref. 28), 144 patients with melanoma from the MEL DFCI 2019 cohort (Supplementary Table S4; ref. 29), 110 patients with melanoma from the skin cutaneous melanoma (SKCM) DFCI 2015 cohort (Supplementary Table S5; ref. 30), 261 patients with clear-cell renal cell carcinoma (ccRCC) from the ccRCC DFCI 2020 cohort (Supplementary Table S6; ref. 31), 92 patients with gastrointestinal cancer (GIC) from the GIC PUCH 2021 cohort (Supplementary Table S7; ref. 32), and 74 patients with non–small cell lung cancer (NSCLC) from the NSCLC MSKCC 2018 cohort (Supplementary Table S8; ref. 33). Patients’ clinical response to ICI therapy was assessed by RECIST (v1.1). Durable clinical benefit was defined as complete response, partial response, or stable disease lasting longer than 6 months; all other patients were classified as having no durable benefit.

Mutation data from 9,433 patients with primary tumors from The Cancer Genome Atlas (TCGA) Research Network were collected to evaluate the frequency of SWI/SNF complex gene alterations.

Recognition of SWI/SNF complex genes

We recognized 31 SWI/SNF genes according to the HUGO Gene Nomenclature Committee at the European Bioinformatics Institute (gene group: SWI/SNF related BAF chromatin remodeling complexes; https://www.genenames.org/data/genegroup/#!/group/2091).

Mutation analysis

In this study, only nonsynonymous mutations were included (the Variant Classification includes Frame_Shift_Del, Frame_Shift_Ins, In_Frame_Del, In_Frame_Ins, Missense_Mutation, Nonsense_Mutation, Nonstop_Mutation, Splice_Site, and Translation_Start_Site). We grouped the patients into those with no mutations in the 31 SWI/SNF genes (unaltered group) and those with mutations in one or more SWI/SNF gene (altered group).

Curation of published genetic alteration biomarkers

We collected the following previously published genetic alteration biomarkers associated with response to ICI therapy: TMB (34), SERPINB3/SERPINB4 mutations (35), receptor tyrosine kinase superfamily mutations (EGFR, ERBB2, MET, FGFR1, and IGF1R; ref. 36), DNA damage response pathway mutations (BRCA1, BRCA2, ATM, POLE, ERCC2, FANCA, MSH2, MLH1, POLD1, and MSH6; ref. 37), JAK1/JAK2 mutations (38), TET1 mutations (39), MUC16 mutations (40), BAP1 mutations (41), KRAS mutations (42), TP53 mutations (42), EPHA7 mutations (43), STK11 mutations (42), B2M mutations (44), and PTEN mutations (45, 46).

Statistical analysis

We compared Kaplan–Meier survival curves using the log-rank test [ggsurvplot() function, from “survminer” package] and computed HRs between groups using univariable Cox regression analysis [coxph() function, from “survival” package]. Additional multivariate Cox regression including SWI/SNF complex alteration, TMB (47), and other available clinical characteristics was performed to confirm the association of SWI/SNF complex alteration with OS [coxph() function, from “survival” package]. Comparisons of clinical response between two groups were examined using the Mann–Whitney test [wilcox.test() function, one-sided, from “stats” package]. Fisher exact test was used to compare proportions of durable clinical benefit between two groups [fisher.test() function, one-sided, from “stats” package]. We used the random forest method to measure variable importance and perform variable selection via the “randomforestSRC” package (48). We use the “ctree” function with default parameters from the “party” package to build the optimal survival trees (49). Statistical analysis and data visualization were performed using the R software (version 4.1.1). Analysis methods for each cohort are depicted in Supplementary Fig. S1.

Data availability statement

The data analyzed in this study were obtained from TCGA PanCanAtlas (https://gdc.cancer.gov/about-data/publications/pancanatlas), cBioPortal at tmb_mskcc_2018 (https://www.cbioportal.org/study/summary?id=tmb_mskcc_2018), mixed_allen_2018 (https://www.cbioportal.org/study/summary?id=mixed_allen_2018), mel_dfci_2019 (https://www.cbioportal.org/study/summary?id=mel_dfci_2019), skcm_dfci_2015 (https://www.cbioportal.org/study/summary?id=skcm_dfci_2015), and nsclc_mskcc_2018 (https://www.cbioportal.org/study/summary?id=nsclc_mskcc_2018; ref. 50). The other data analyzed in this study (ccRCC DFCI 2020 and GIC PUCH 2021) were obtained from the Supplementary Data of the original reports (31, 32). Any new data generated in the study are available within the article and its Supplementary Data or from the corresponding author upon reasonable request.

Ethics approval

The current study investigated the publicly available data from TCGA, cBioPortal, and publications described in Data availability statement. The ethical approval was waived by Fujian Cancer Hospital.

The prevalence of SWI/SNF complex genes alteration in cancers

The SWI/SNF complex includes three large macromolecular complexes—BRG1/BRM-associated factor (BAF), also known as canonical BAF; polybromo-associated BAF (PBAF), and noncanonical BAF (also termed GBAF) complexes—which are collectively composed of 31 proteins. The shared subunits of the three complex members and their unique subunits are shown in Fig. 1A. A total of 9,433 patients from TCGA's Pan-Cancer Atlas who underwent WES to detect gene alterations were included in the initial analysis. Among these patients there was an overall frequency of 27.2% (2,563 patients) carrying SWI/SNF complex protein-coding gene variants. Across different cancer types, the mutational frequency of ARID1A (8.3%) was the highest, followed by PBRM1 (4.2%), ARID2 (4.1%), SMARCA4 (4.0%), ARID1B (3.4%), SMARCA2 (2.7%), BCL11A (2.2%), and SMARCC2 (2.0%; Supplementary Table S9). Nonsynonymous mutations in SWI/SNF genes were most highly enriched in uterine corpus endometrial carcinoma (59.2%), followed by SKCM (56.6%), bladder urothelial carcinoma (56.0%), ccRCC (TCGA code: KIRC; 48.9%), and stomach adenocarcinoma (42.7%; Supplementary Table S10). The heat map in Fig. 1B depicts the frequency of nonsynonymous mutations in the 31 genes encoding components of the SWI/SNF complexes across cancer types (Supplementary Table S11).

Treatment outcomes of patients with SWI/SNF complex alterations using targeted sequencing

As shown in the Venn diagram, genes encoding seven of the 31 SWI/SNF complex proteins were included in the MSK-IMPACT gene panel, including ARID1A, ARID1B, ARID2, PBRM1, SMARCA4, SMARCB1, and SMARCD1 (Fig. 2A). Grouping all patients from TMB MSKCC 2018 according to the mutation status of these seven genes, a survival analysis showed that mutation status was significantly associated with OS benefit [28.0 vs. 16.0 months; HR, 0.73 (95% confidence interval, CI, 0.62–0.85), log-rank P < 0.0001] (Fig. 2B). Consistent with a previous report (9), there was no association of SWI/SNF complex alterations with OS upon adjustment for TMB in a multivariate Cox model [Supplementary Table S12; HR 1 (95% CI, 0.85–1.22), P = 0.856 (sex, age, drug type, cancer type, and TMB)]. Furthermore, we performed survival analyses for each cancer type to determine whether there were differences between the two groups. With a trend toward better survival, the median OS of the altered groups was better than that of the unaltered groups. However, the difference in survival was only statistically significant in head and neck squamous cell carcinoma [median OS 28 vs. 10 months; HR, 0.44 (95% CI, 0.22–0.88), log-rank P = 0.016] (Fig. 2C). There was no significant difference in survival in melanoma (median OS 58 vs. 41 months; HR, 0.73 (95% CI, 0.48–1.12), log-rank P = 0.15], colorectal cancer [median OS 31 vs. 13 months; HR, 0.56 (95% CI, 0.29–1.08), log-rank P = 0.081], or breast cancer [median OS not reached vs. 5 months; HR, 0.31 (95% CI, 0.073–1.32), log-rank P = 0.088] (Fig. 2DF).

Treatment outcomes of patients with melanoma with SWI/SNF complex alteration under WES

We comprehensively profiled the mutational landscape of SWI/SNF complex genes in three WES melanoma cohorts (the Mixed Allen 2018 cohort, the MEL DFCI 2019 cohort, and the SKCM DFCI 2015 cohort) and the combination of these cohorts. The most frequently altered genes in the combined cohort were ARID2 (12.1%), ARID1B (8.4%), SMARCA4 (8.1%), ARID1A (7.9%), and BICRA (6.7%), with some differences in the most frequently altered genes among the three individual cohorts (Fig. 3A). In the combined cohort, survival analysis showed that alterations in any one or more of the 31 genes encoding the SWI/SNF complex were significantly associated with improved OS (median OS 33.5 vs. 11.1 months, log-rank P < 0.0001; Fig. 3B). When TMB with a binary cutoff was included in multivariate Cox regression, SWI/SNF complex alterations still showed a significant correlation with OS benefit [HR, 0.63 (95% CI, 0.47–0.85), P = 0.003] (Fig. 3C). Similar results were obtained when TMB was included in multivariate analysis as a continuous variable (Supplementary Fig. S2A). In addition, the result of regression analysis showed that TMB increased with the number of altered genes in the SWI/SNF complex (Fig. 3D). And the clinical response of the altered group was significantly better than that of the unaltered group (P = 0.002; Supplementary Fig. S3A).

Consistent with the analysis of the combined cohort, the median OS of the altered group was significantly better than that of the unaltered group in the Mixed Allen 2018 [median OS 34.5 vs. 11.6 months; HR, 0.61 (95% CI, 0.39–0.94), log-rank P = 0.025; Padjusted = 0.038 (age, sex, TMB)] (Fig. 3E; Supplementary Fig. S2B) and the MEL DFCI 2019 [median OS not reach vs. 17.4 months; HR, 0.50 (95% CI, 0.31–0.81), log-rank P = 0.004; Padjusted = 0.021 (sex, TMB)] (Fig. 3E; Supplementary Fig. S2B). However, it was not significant in the SKCM DFCI 2015 cohort [median OS 15.03 vs. 8.12 months; HR, 0.73 (95% CI, 0.46–1.15), log-rank P = 0.17; Padjusted = 0.344 (age, sex, TMB)] (Fig. 3E; Supplementary Fig. S2B).

Construction of a SWI/SNF signature

We used the random survival forest method to measure the variable importance in the Mixed Allen 2018 cohort. A large positive variable importance indicates a high predictive ability, whereas zero or negative values indicate noisy variables (48). We defined 14 genes with positive variable importance as a SWI/SNF signature (Fig. 4A). We evaluated the performance of the SWI/SNF signature as a survival biomarker in the other two melanoma cohorts (the MEL DFCI 2019 cohort and the SKCM DFCI 2015 cohort) and the combination of the three melanoma cohorts. We found that alterations in the SWI/SNF signature were associated with significant clinical benefits in all cohorts (all log-rank P < 0.05; Fig. 4B).

Furthermore, stratification analysis in the combined cohort revealed that favorable clinical outcomes for the signature-altered group versus the signature-unaltered group were prominent and consistent across subgroups of sex and TMB levels (Fig. 5A). We divided the patients into four groups based on signature status and sex or TMB level and compared their Kaplan–Meier curves. Consistent with the stratification analysis, the signature-altered group achieved longer OS regardless of sex or TMB level (Fig. 5B and C). In TMB-high patients, the altered status of the SWI/SNF signature effectively identified patients with better prognosis.

Next, we assessed the importance of sex and genetic alteration biomarkers, including our SWI/SNF signature, by using the random survival forest method in the combined cohort. We found that the SWI/SNF signature was the most important among various previously reported biomarkers, followed by SERPINB3/SERPINB4 (Fig. 6A). Considering all the above features, the SWI/SNF signature and SERPINB3/SERPINB4 were selected to build the optimal survival tree (Fig. 6B). Comparing the predictive ability of genetic alteration biomarkers, it was found that the SWI/SNF signature combined with SERPINB3/SERPINB4 had the best predictive ability, with an AUC of 0.683 (Fig. 6C).

SWI/SNF complex alterations and clinical outcomes association in cancers other than melanoma

To determine whether the altered SWI/SNF complex can be used as a biomarker for immunotherapy in other cancers, we compared the prognostic differences of all 31 genes encoding the SWI/SNF complex proteins altered and unaltered groups in the ccRCC DFCI 2020 cohort, the GIC PUCH 2021 cohort, and the NSCLC MSKCC 2018 cohort. We found that the SWI/SNF complex altered group had significantly improved OS compared with that of the unaltered group in ccRCC [Fig. 7A, median OS 27.9 vs. 20.7 months; HR, 0.64 (95% CI, 0.48–0.86), log-rank P = 0.003] and GIC [Fig. 7B, median OS 19.20 vs. 7.43 months; HR, 0.39 (95% CI, 0.19–0.78), log-rank P = 0.006]. In NSCLC, the median PFS of the SWI/SNF complex altered group was also better than that of the unaltered group, but the difference was not significant [Fig. 7C, median PFS 9.8 vs. 5.2 months; HR, 0.60 (95% CI, 0.33–1.07), log-rank P = 0.079]. When TMB with a binary cutoff was included in multivariate analysis, all 31 genes of SWI/SNF complex alterations still showed the prognosis value of SWI/SNF genomic alteration in ccRCC [Fig. 7D, HR, 0.62 (95% CI, 0.46–0.85), P = 0.003] and GIC [Fig. 7E, HR, 0.42 (95% CI, 0.18–1.01), P = 0.053], but not in NSCLC [Fig. 7F, HR, 0.86 (95% CI, 0.44–1.70), P = 0.667]. Similar results were obtained when TMB was included in multivariate analysis as a continuous variable (Supplementary Fig. S4). In these tumors, clinical responses between the two groups were compared. In the GIC PUCH 2021 cohort and the NSCLC MSKCC 2018 cohort, significantly more SWI/SNF altered patients got durable clinical benefit than the unaltered patients (Supplementary Fig. S3C and S3D, P = 0.004 and 0.048, respectively). In the ccRCC DFCI 2020 cohort and the NSCLC MSKCC 2018 cohort, the clinical outcomes of SWI/SNF altered groups were better than that of unaltered groups, although the differences were not significant (P = 0.264 and 0.208, respectively; Supplementary Fig. S3B and S3D).

Furthermore, we compared the median OS or PFS of the signature-altered group with the signature-unaltered group in these three cancer cohorts, and the differences were all significant (all log-rank P < 0.05; Fig. 7GI). Stratification analysis showed that the association of SWI/SNF signature with survival is consistent across age, sex, and mutation levels (Supplementary Fig. S5).

Unlike prior studies focused on either an individual gene or a predefined set of SWI/SNF genes, we evaluated the association between alterations in all 31 SWI/SNF genes and clinical outcomes for ICI therapy using WES data from six patient cohorts. Through univariate analysis, we found that ICI-treated patients with complex alterations had better prognosis and clinical response. Additional multivariate Cox regression including SWI/SNF complex alteration, TMB, and other available clinical characteristics confirmed the prognostic value of SWI/SNF genomic alteration in melanoma, ccRCC, and GIC. Furthermore, we used the random forest method for variable screening to obtain 14 genes as a SWI/SNF signature for potential clinical application.

Within contrast to previous studies that included loss-of-function mutations—defined as nonsense mutations, frame-shift insertions or deletions, splice-site variants affecting consensus nucleotides, or homozygous deletions in the analysis excluding missense mutations (9, 18, 51)—we analyzed all nonsynonymous mutations in all 31 of the SWI/SNF genes. Although we could not confidently assess the functional outcomes of missense mutations, a large sample size study revealed the functional characterization of these hotspot missense mutations within the helicase domain of SMARCA4, a core subunit of the SWI/SNF complex. This study found that missense mutations in SMARCA4 markedly reduced remodeling activity due to a deficiency in opening chromatin and inducing gene expression (52). In addition, SWI/SNF complex alterations are frequently observed in a broad spectrum of human cancers, occurring in 27.2% of all cancers, while most other genes mutated at such high frequencies have been studied for many decades. The oncogenic effects of SWI/SNF gene mutations have only recently been recognized; therefore, relatively little is known about these changes and the tumorigenic effect of missense mutations cannot be ruled out. Consequently, we included all the missense mutations in our analysis.

To the best of our knowledge, this is the first analysis of the association between all 31 members of the SWI/SNF complex alterations and clinical outcomes following ICI therapy across multiple cancer types, suggesting that SWI/SNF genomic alterations can be considered biomarkers of response to ICI therapy in multiple cancers. Emerging evidence has demonstrated that the SWI/SNF complex plays a role in modulation of antitumor immunity. The alteration of SWI/SNF genes in some cancer types is associated with enhanced CD8+ T-cell infiltration, increased sensitivity to T cell–mediated cytotoxicity, and improved responses to ICI therapy (8). For example, Pan and colleagues demonstrated that patients with RCC may benefit from ICI therapy when SWI/SNF alterations are present in the tumors. Because the expression of immune-stimulatory gene sets associated with TNFα signaling and IL6–JAK–STAT3 signaling is enhanced, sensitization to T cell–mediated killing is reinforced and accessibility to IFNγ-inducible genes is increased, creating an immune-responsive environment (11). Recently, Belk and colleagues systematically dissected the genetic regulators of T-cell exhaustion through in vitro and in vivo CRISPR/Cas9 screens. They found that depletion of the cBAF complex subunit ARID1A improved T-cell function and reduced the transcriptional and epigenetic hallmarks of exhaustion, which improved antitumor immunity after adoptive T-cell transfer, suggesting that SWI/SNF complex genomic alterations may contribute to more active immunotherapeutic responsiveness (53). In addition, Conway and colleagues found that PBAF complex mutations were significantly associated with improved OS and PFS when patients with (N)RAS melanoma were treated with immunotherapy (17). These mechanistic studies and clinical studies all support our conclusions.

In addition, we observed a positive correlation between the number of SWI/SNF complex altered genes and TMB, while SWI/SNF complex alterations could still be an independent prognostic factor for ICI therapy. This SWI/SNF complex–related benefit of ICI therapy may be associated with the role of the SWI/SNF complex in transcriptional regulation and DNA repair. The SWI/SNF complex mediates immune responses through transcriptional regulation (8, 11). Multiple components of SWI/SNF complex have been reported to be involved in DNA damage repair mechanisms (54–56). Accordingly, we speculate that SWI/SNF complex altered tumors may have more mutations and neoantigens due to DNA damage repair defects, so they are more sensitive to ICI treatment. However, the specific mechanism of SWI/SNF complex-related immunotherapy benefits needs to be further determined by experimental models.

We used the random forest method for variable screening and identified 14 genes as a SWI/SNF signature. Significant correlations were observed between SWI/SNF signature alterations and improved OS in all cohorts, including melanoma, ccRCC, GIC, and NSCLC, regardless of age, sex, or TMB. These results provide evidence that SWI/SNF signature alterations can be regarded as a biomarker of response to ICI therapy in multiple cancers.

However, this study had several limitations that should be cautiously interpreted. First, the number of cohorts reporting clinical outcomes following ICI therapy for which there are WES data was limited. Second, our data were curated from previous studies, and prospective controlled trials should be conducted for further validation, using comprehensive baseline data. Nevertheless, our study has implications for the clinical evaluation of the efficacy of ICI therapy in that we should consider SWI/SNF genomic alteration as a promising biomarker.

No disclosures were reported.

D. Wang: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, investigation, visualization, methodology, writing–original draft, writing–review and editing. J. Wang: Supervision, validation, investigation, writing–review and editing. D. Zhou: Data curation, methodology, writing–review and editing. Z. Wu: Data curation, software, validation, writing–review and editing. W. Liu: Data curation, validation, writing–review and editing. Y. Chen: Supervision, validation, writing–review and editing. G. Chen: Conceptualization, supervision, investigation, methodology, project administration, writing–review and editing. J. Zhang: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, validation, investigation, writing–original draft, project administration, writing–review and editing.

This study was funded by Scientific Research Foundation of Fujian Cancer Hospital (grant no.: 2021YNY01) for J. Zhang; the work was also partially supported by Scientific Research Foundation of Fujian Cancer Hospital (grant no.: YJ-YJ-02) for D. Wang.

The results published here are in whole or part based upon data generated by TCGA Research Network: https://www.cancer.gov/tcga.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Immunology Research Online (http://cancerimmunolres.aacrjournals.org/).

1.
Abril-Rodriguez
G
,
Ribas
A
.
SnapShot: immune checkpoint inhibitors
.
Cancer Cell
2017
;
31
:
848
.
2.
Twomey
JD
,
Zhang
B
.
Cancer immunotherapy update: FDA-approved checkpoint inhibitors and companion diagnostics
.
AAPS J
2021
;
23
:
39
.
3.
Sharma
P
,
Siddiqui
BA
,
Anandhan
S
,
Yadav
SS
,
Subudhi
SK
,
Gao
J
, et al
.
The next decade of immune checkpoint therapy
.
Cancer Discov
2021
;
11
:
838
57
.
4.
Marcus
L
,
Fashoyin-Aje
LA
,
Donoghue
M
,
Yuan
M
,
Rodriguez
L
,
Gallagher
PS
, et al
.
FDA approval summary: pembrolizumab for the treatment of tumor mutational burden–high solid tumors
.
Clin Cancer Res
2021
;
27
:
4685
9
.
5.
Fang
H
,
Bertl
J
,
Zhu
X
,
Lam
TC
,
Wu
S
,
Shih
DJH
, et al
.
Tumour mutational burden is overestimated by target cancer gene panels
.
J Natl Cancer Cent
2023
;
3
:
56
64
.
6.
Strickler
JH
,
Hanks
BA
,
Khasraw
M
.
Tumor mutational burden as a predictor of immunotherapy response: is more always better?
Clin Cancer Res
2021
;
27
:
1236
41
.
7.
Nassar
AH
,
Adib
E
,
Abou Alaiwi
S
,
El Zarif
T
,
Groha
S
,
Akl
EW
, et al
.
Ancestry-driven recalibration of tumor mutational burden and disparate clinical outcomes in response to immune checkpoint inhibitors
.
Cancer Cell
2022
;
40
:
1161
72
.
8.
Zhou
M
,
Yuan
J
,
Deng
Y
,
Fan
X
,
Shen
J
.
Emerging role of SWI/SNF complex deficiency as a target of immune checkpoint blockade in human cancers
.
Oncogenesis
2021
;
10
:
3
.
9.
Abou Alaiwi
S
,
Nassar
AH
,
Xie
W
,
Bakouny
Z
,
Berchuck
JE
,
Braun
DA
, et al
.
Mammalian SWI/SNF complex genomic alterations and immune checkpoint blockade in solid tumors
.
Cancer Immunol Res
2020
;
8
:
1075
84
.
10.
Shen
J
,
Ju
Z
,
Zhao
W
,
Wang
L
,
Peng
Y
,
Ge
Z
, et al
.
ARID1A deficiency promotes mutability and potentiates therapeutic antitumor immunity unleashed by immune checkpoint blockade
.
Nat Med
2018
;
24
:
556
62
.
11.
Pan
D
,
Kobayashi
A
,
Jiang
P
,
Ferrari de Andrade
L
,
Tay
RE
,
Luoma
AM
, et al
.
A major chromatin regulator determines resistance of tumor cells to T cell-mediated killing
.
Science
2018
;
359
:
770
5
.
12.
Leruste
A
,
Tosello
J
,
Ramos
RN
,
Tauziède-Espariat
A
,
Brohard
S
,
Han
Z-Y
, et al
.
Clonally expanded T cells reveal immunogenicity of rhabdoid tumors
.
Cancer Cell
2019
;
36
:
597
612
.
13.
Goswami
S
,
Chen
Y
,
Anandhan
S
,
Szabo
PM
,
Basu
S
,
Blando
JM
, et al
.
ARID1A mutation plus CXCL13 expression act as combinatorial biomarkers to predict responses to immune checkpoint therapy in mUCC
.
Sci Transl Med
2020
;
12
:
eabc4220
.
14.
Liu
X-D
,
Kong
W
,
Peterson
CB
,
McGrail
DJ
,
Hoang
A
,
Zhang
X
, et al
.
PBRM1 loss defines a nonimmunogenic tumor phenotype associated with checkpoint inhibitor resistance in renal carcinoma
.
Nat Commun
2020
;
11
:
2135
.
15.
Li
J
,
Wang
W
,
Zhang
Y
,
Cieślik
M
,
Guo
J
,
Tan
M
, et al
.
Epigenetic driver mutations in ARID1A shape cancer immune phenotype and immunotherapy
.
J Clin Invest
2020
;
130
:
2712
26
.
16.
Botta
GP
,
Kato
S
,
Patel
H
,
Fanta
P
,
Lee
S
,
Okamura
R
, et al
.
SWI/SNF complex alterations as a biomarker of immunotherapy efficacy in pancreatic cancer
.
JCI Insight
2021
;
6
:
e150453
.
17.
Conway
JR
,
Dietlein
F
,
Taylor-Weiner
A
,
AlDubayan
S
,
Vokes
N
,
Keenan
T
, et al
.
Integrated molecular drivers coordinate biological and clinical states in melanoma
.
Nat Genet
2020
;
52
:
1373
83
.
18.
Miao
D
,
Margolis
CA
,
Gao
W
,
Voss
MH
,
Li
W
,
Martini
DJ
, et al
.
Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma
.
Science
2018
;
359
:
801
6
.
19.
Braun
DA
,
Ishii
Y
,
Walsh
AM
,
Van Allen
EM
,
Wu
CJ
,
Shukla
SA
, et al
.
Clinical validation of PBRM1 alterations as a marker of immune checkpoint inhibitor response in renal cell carcinoma
.
JAMA Oncol
2019
;
5
:
1631
3
.
20.
McDermott
DF
,
Huseni
MA
,
Atkins
MB
,
Motzer
RJ
,
Rini
BI
,
Escudier
B
, et al
.
Clinical activity and molecular correlates of response to atezolizumab alone or in combination with bevacizumab versus sunitinib in renal cell carcinoma
.
Nat Med
2018
;
24
:
749
57
.
21.
Motzer
RJ
,
Robbins
PB
,
Powles
T
,
Albiges
L
,
Haanen
JB
,
Larkin
J
, et al
.
Avelumab plus axitinib versus sunitinib in advanced renal cell carcinoma: biomarker analysis of the phase 3 JAVELIN Renal 101 trial
.
Nat Med
2020
;
26
:
1733
41
.
22.
Wang
N
,
Qin
Y
,
Du
F
,
Wang
X
,
Song
C
.
Prevalence of SWI/SNF genomic alterations in cancer and association with the response to immune checkpoint inhibitors: a systematic review and meta-analysis
.
Gene
2022
;
834
:
146638
.
23.
Hakimi
AA
,
Attalla
K
,
DiNatale
RG
,
Ostrovnaya
I
,
Flynn
J
,
Blum
KA
, et al
.
A pan-cancer analysis of PBAF complex mutations and their association with immunotherapy response
.
Nat Commun
2020
;
11
:
4168
.
24.
Courtet
K
,
Laizet
Y
,
Lucchesi
C
,
Bessede
A
,
Italiano
A
.
Inactivating mutations in genes encoding for components of the BAF/PBAF complex and immune-checkpoint inhibitor outcome
.
Biomark Res
2020
;
8
:
26
.
25.
Li
L
,
Li
M
,
Jiang
Z
,
Wang
X
.
ARID1A mutations are associated with increased immune activity in gastrointestinal cancer
.
Cells
2019
;
8
:
678
.
26.
Zhu
G
,
Shi
R
,
Li
Y
,
Zhang
Z
,
Xu
S
,
Chen
C
, et al
.
ARID1A, ARID1B, and ARID2 mutations serve as potential biomarkers for immune checkpoint blockade in patients with non-small cell lung cancer
.
Front Immunol
2021
;
12
:
670040
.
27.
Samstein
RM
,
Lee
CH
,
Shoushtari
AN
,
Hellmann
MD
,
Shen
R
,
Janjigian
YY
, et al
.
Tumor mutational load predicts survival after immunotherapy across multiple cancer types
.
Nat Genet
2019
;
51
:
202
6
.
28.
Miao
D
,
Margolis
CA
,
Vokes
NI
,
Liu
D
,
Taylor-Weiner
A
,
Wankowicz
SM
, et al
.
Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors
.
Nat Genet
2018
;
50
:
1271
81
.
29.
Liu
D
,
Schilling
B
,
Liu
D
,
Sucker
A
,
Livingstone
E
,
Jerby-Arnon
L
, et al
.
Integrative molecular and clinical modeling of clinical outcomes to PD1 blockade in patients with metastatic melanoma
.
Nat Med
2019
;
25
:
1916
27
.
30.
Van Allen
EM
,
Miao
D
,
Schilling
B
,
Shukla
SA
,
Blank
C
,
Zimmer
L
, et al
.
Genomic correlates of response to CTLA-4 blockade in metastatic melanoma
.
Science
2015
;
350
:
207
11
.
31.
Braun
DA
,
Hou
Y
,
Bakouny
Z
,
Ficial
M
,
Sant' Angelo
M
,
Forman
J
, et al
.
Interplay of somatic alterations and immune infiltration modulates response to PD-1 blockade in advanced clear cell renal cell carcinoma
.
Nat Med
2020
;
26
:
909
18
.
32.
Jiao
X
,
Wei
X
,
Li
S
,
Liu
C
,
Chen
H
,
Gong
J
, et al
.
A genomic mutation signature predicts the clinical outcomes of immunotherapy and characterizes immunophenotypes in gastrointestinal cancer
.
NPJ Precis Oncol
2021
;
5
:
36
.
33.
Hellmann
MD
,
Nathanson
T
,
Rizvi
H
,
Creelan
BC
,
Sanchez-Vega
F
,
Ahuja
A
, et al
.
Genomic features of response to combination immunotherapy in patients with advanced non-small-cell lung cancer
.
Cancer Cell
2018
;
33
:
843
52
.
34.
Rizvi
NA
,
Hellmann
MD
,
Snyder
A
,
Kvistborg
P
,
Makarov
V
,
Havel
JJ
, et al
.
Mutational landscape determines sensitivity to PD-1 blockade in non–small cell lung cancer
.
Science
2015
;
348
:
124
8
.
35.
Riaz
N
,
Havel
JJ
,
Kendall
SM
,
Makarov
V
,
Walsh
LA
,
Desrichard
A
, et al
.
Recurrent SERPINB3 and SERPINB4 mutations in patients who respond to anti-CTLA4 immunotherapy
.
Nat Genet
2016
;
48
:
1327
9
.
36.
Anagnostou
V
,
Niknafs
N
,
Marrone
K
,
Bruhm
DC
,
White
JR
,
Naidoo
J
, et al
.
Multimodal genomic features predict outcome of immune checkpoint blockade in non-small-cell lung cancer
.
Nat Cancer
2020
;
1
:
99
111
.
37.
Conway
JR
,
Kofman
E
,
Mo
SS
,
Elmarakeby
H
,
Van Allen
E
.
Genomics of response to immune checkpoint therapies for cancer: implications for precision medicine
.
Genome Med
2018
;
10
:
93
.
38.
Shin
DS
,
Zaretsky
JM
,
Escuin-Ordinas
H
,
Garcia-Diaz
A
,
Hu-Lieskovan
S
,
Kalbasi
A
, et al
.
Primary resistance to PD-1 blockade mediated by JAK1/2 mutations
.
Cancer Discov
2017
;
7
:
188
201
.
39.
Wu
H-X
,
Chen
Y-X
,
Wang
Z-X
,
Zhao
Q
,
He
M-M
,
Wang
Y-N
, et al
.
Alteration in TET1 as potential biomarker for immune checkpoint blockade in multiple cancers
.
J Immunother Cancer
2019
;
7
:
264
.
40.
Zhang
L
,
Han
X
,
Shi
Y
.
Association of MUC16 mutation with response to immune checkpoint inhibitors in solid tumors
.
JAMA Network Open
2020
;
3
:
e2013201
.
41.
Shrestha
R
,
Nabavi
N
,
Lin
Y-Y
,
Mo
F
,
Anderson
S
,
Volik
S
, et al
.
BAP1 haploinsufficiency predicts a distinct immunogenic class of malignant peritoneal mesothelioma
.
Genome Med
2019
;
11
:
8
.
42.
Aredo
JV
,
Padda
SK
,
Kunder
CA
,
Han
SS
,
Neal
JW
,
Shrager
JB
, et al
.
Impact of KRAS mutation subtype and concurrent pathogenic mutations on non-small cell lung cancer outcomes
.
Lung Cancer
2019
;
133
:
144
50
.
43.
Zhang
Z
,
Wu
H-X
,
Lin
W-H
,
Wang
Z-X
,
Yang
L-P
,
Zeng
Z-L
, et al
.
EPHA7 mutation as a predictive biomarker for immune checkpoint inhibitors in multiple cancers
.
BMC Med
2021
;
19
:
26
.
44.
Gettinger
S
,
Choi
J
,
Hastings
K
,
Truini
A
,
Datar
I
,
Sowell
R
, et al
.
Impaired HLA class I antigen processing and presentation as a mechanism of acquired resistance to immune checkpoint inhibitors in lung cancer
.
Cancer Discov
2017
;
7
:
1420
35
.
45.
Peng
W
,
Chen
JQ
,
Liu
C
,
Malu
S
,
Creasy
C
,
Tetzlaff
MT
, et al
.
Loss of PTEN promotes resistance to T cell–mediated immunotherapy
.
Cancer Discov
2016
;
6
:
202
16
.
46.
Litchfield
K
,
Reading
JL
,
Puttick
C
,
Thakkar
K
,
Abbosh
C
,
Bentham
R
, et al
.
Meta-analysis of tumor- and T cell-intrinsic mechanisms of sensitization to checkpoint inhibition
.
Cell
2021
;
184
:
596
614
.
47.
Zheng
M
.
Tumor mutation burden for predicting immune checkpoint blockade response: the more, the better
.
J Immunother Cancer
2022
;
10
:
e003087
.
48.
Hemant
I
,
Udaya
BK
,
Eugene
HB
,
Michael
SL
.
Random survival forests
.
Ann Appl Stat
2008
;
2
:
841
60
.
49.
Hothorn
T
,
Hornik
K
,
Zeileis
A
.
Unbiased recursive partitioning: a conditional inference framework
.
J Comput Graph Statist
2006
;
15
:
651
74
.
50.
Gao
J
,
Aksoy
BA
,
Dogrusoz
U
,
Dresdner
G
,
Gross
B
,
Sumer
SO
, et al
.
Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal
.
Sci Signal
2013
;
6
:
pl1
.
51.
Huang
HT
,
Chen
SM
,
Pan
LB
,
Yao
J
,
Ma
HT
.
Loss of function of SWI/SNF chromatin remodeling genes leads to genome instability of human lung cancer
.
Oncol Rep
2015
;
33
:
283
91
.
52.
Fernando
TM
,
Piskol
R
,
Bainer
R
,
Sokol
ES
,
Trabucco
SE
,
Zhang
Q
, et al
.
Functional characterization of SMARCA4 variants identified by targeted exome-sequencing of 131,668 cancer patients
.
Nat Commun
2020
;
11
:
5551
.
53.
Belk
JA
,
Yao
W
,
Ly
N
,
Freitas
KA
,
Chen
YT
,
Shi
Q
, et al
.
Genome-wide CRISPR screens of T cell exhaustion identify chromatin remodeling factors that limit T cell persistence
.
Cancer Cell
2022
;
40
:
768
86
.
54.
Mathur
R
,
Roberts
CWM
.
SWI/SNF (BAF) complexes: guardians of the epigenome
.
Ann Rev Cancer Biol
2018
;
2
:
413
27
.
55.
Mittal
P
,
Roberts
CWM
.
The SWI/SNF complex in cancer — biology, biomarkers and therapy
.
Nat Rev Clin Oncol
2020
;
17
:
435
48
.
56.
Centore
RC
,
Sandoval
GJ
,
Soares
LMM
,
Kadoch
C
,
Chan
HM
.
Mammalian SWI/SNF chromatin remodeling complexes: emerging mechanisms and therapeutic strategies
.
Trends Genet
2020
;
36
:
936
50
.
This open access article is distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license.