Tumor Mutation Burden Level Research Articles

Clinical significance of tumor mutation burden and DNA damage repair in advanced stage non-small cell lung cancer patients.

This study aimed to investigate the impact of tumor mutational burden (TMB) and DNA damage repair (DDR) gene alteration on overall survival (OS) in advanced non-small cell lung cancer (NSCLC) patients. A DNA library of cancer cells from 67 NSCLC patients in stages III-IV was constructed for next-generation sequencing (NGS). Geneseeq422 probes were used for hybridization enrichment. The target-enriched library was sequenced on HiSeqNGS platforms, and we analyzed the relevant signaling pathways. Then, we correlated the OS of the patients with TMB and DDR mutations. Many significant alterations were found, including in the EGFR, p53, KRAS, RB1, ERBB2, NF1, DNMT3A, ALK, MYC, PIK3CA, ROS1, BRAF, ARID1A, PTEN, CDKN2A, and FGF19 genes. We also identified many mutations in the genes relevant to the DDR pathway. Interestingly, we found that the TMB of patients with DDR gene mutations was dramatically higher than that in the DDR wild-type (WT). Univariable analysis showed that DNMT3A, RB1, DDR pathway-related gene mutations, and TMB were critical factors for the effects on OS. Multivariable analysis confirmed that DNMT3A and mutations in the DDR pathway-related genes were important for predicting OS. Multiple mutations in the genes of the DDR pathway caused higher TMB levels, which resulted in longer OS. By contrast, OS was significantly longer in patients with non-DNMT3A mutations than in those with DNMT3A variants. DNMT3A alteration in NSCLC patients led to poor outcomes.

Tumor mutation burden associated with miRNA-gene interaction outcome mediates the survival of patients with liver hepatocellular carcinoma.

Tumor mutation burden (TMB) is associated with immunogenic responses and the survival of cancer patients. This study demonstrates how TMB levels impact the immune-related cells, genes, and miRNAs, and how miRNA/gene interactions respond to variations in the survival rate of patients with liver hepatocellular carcinoma (LIHC). LIHC patients were divided into two groups, either a low TMB (< median) or a high TMB (≥ median) group. We found that high TMB plays a positive role in immune-mediated infiltration, generating more CD4 T-cells and memory B cells. Among the 21 immune genes that altered significantly, only C9orf24 and CYP1A1 were expected to up-regulate in LIHC patients with high TMB. A total of 19 miRNAs, which regulate various functional pathways, were significantly altered in patients with LIHC. One of the miRNA/gene pair, hsa-miR-33a/ALDH1A3 was significantly associated with the survival rate of LIHC patients. Our results suggest that LIHC patients with high TMB can be treated more effectively with immunotherapy.

Abstract D088: Analysis of tumor mutational burden, cancer immunotherapy use and outcomes based on genomic ancestry in non-small cell lung cancer patients treated in community practice in the United States

Abstract Introduction Racial disparities in cancer care and outcome are multifactorial and well documented, being attributed to social determinants, access to and quality of care. In addition, variation in biologic drivers by self-reported race/ethnicity or genetic ancestry may contribute to inequities in outcomes. Differences in tumor mutational burden (TMB), an important predictor of response to cancer immunotherapy (CIT), have been reported in non-small cell lung cancer (NSCLC) patients of different genomic ancestries. The impact of differing TMB-levels on outcomes and underlying cause is unknown, though smoking is associated with increased TMB. This study seeks to describe the receipt of CIT-containing regimens and the distribution and relationship of TMB with overall survival in NSCLC patients across ancestral groups. Methods 3962 patients with advanced NSCLC diagnosed between 1/1/11 and 4/30/19 were selected from the de-identified Flatiron Health-Foundation Medicine (FMI) linked clinicogenomics database. TMB was categorized as high or low (16mutations (mut)/megabase (Mb) cut off). Genomic ancestry was classified as previously described by FMI. CIT use and TMB status was described by ancestry and the association of TMB status with overall survival (OS) from diagnosis was estimated using Cox proportional hazards models. Analyses were further stratified by genomic ancestry (European vs. Non-European) and receipt of CIT at any point after advanced diagnosis date. Results The study population was 72% European ancestry (EUR), 8% African (AFR), 5% American (AMR), and 3% Asian (ASN). Genomic ancestry was largely correlated with self-reported race. The use of CIT-containing regimens was similar across ancestries, from 36% - 45% for those of ASN and AFR ancestry, respectively. TMB levels varied across ancestry groups; patients of AFR ancestry had the highest median TMB level (9.6 mut/Mb, [Interquartile Range: 4.4 – 15.7]), whereas ASN had lowest (3.5 mut/Mb, [2.6, 13.0]). Median TMB for EUR patients was 7.8 mut/Mb [3.5 – 13.9]. Similar trends were observed among patients with a history of smoking, though overall TMB levels were higher. TMB-high patients accounted for 24% of those of AFR, 20% for EUR, and 27% and 5% for AMR and ASN ancestry respectively. TMB-high was associated with better survival in patients who had ever received CIT (0.63, [95% CI: 0.52, 0.77]). These results were consistent when stratified by EUR (0.62 [0.50, 0.77]) and non-EUR ancestry (0.59 [0.34, 1.02]). Conclusions We observed different levels of TMB across ancestral groups. Differences in prevalence of smoking may be a contributing factor. TMB-high was equally predictive of improved OS among patients receiving CIT for those of EUR and non-EUR ancestry. There were no observed differences in receipt of CIT based on genomic ancestry in this population. These results suggest that addressing reported inequitable access to TMB testing among African American NSCLC patients may improve racial/ethnic disparities in NSCLC outcomes. Citation Format: Nicole Richie, Patricia Luhn, Uzor Ogbu, Gerren Wilson, William Capra, Mark Lee, Otis Brawley. Analysis of tumor mutational burden, cancer immunotherapy use and outcomes based on genomic ancestry in non-small cell lung cancer patients treated in community practice in the United States [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr D088.

A pan-cancer analysis of ARID1A as a potential biomarker for immune checkpoint therapy.

3540 Background: AT-rich interactive domain 1A (ARID1A), encoding a subunit of the BAF (SWI/SNF) chromatin remodeling complex, is correlated with the origination and progress of tumor. Previous research on ARID1A gene revealed that ARID1A deficiency was associated with mismatch repair (MMR) and higher tumor mutation burden (TMB) level in cancer, which might cooperate with immune checkpoint blockade therapy. Methods: Next generation sequencing (NGS) data of 10336 pan-cancer patients were obtained from the MSK-IMPACT Clinical Sequencing cohort (MSKCC). NGS data of 15849 pan-cancer patients from Chinese clinical dataset were analyzed to explore the association between ARID1A gene mutation and TMB. TMB was defined as total number of somatic non-synonymous mutations in coding region. 853 advanced NSCLC patients from two independent cohorts (OAK study cohort and POPLAR study cohort) were used to analyze the correlation between ARID1A alteration and the efficacy of immune checkpoint blockade immunotherapies (ICIs). Results: In total, 8.62% (891/10336) of pan-cancer patients in MSKCC harbored ARID1A mutation and 8.47% (3188/37628) in Chinese cohort. In MSKCC cohort, the highest ARID1A mutation frequency tumor type was endometrial cancer (31.64%, 69/218), bladder cancer (26.95%, 114/423) and hepatobiliary cancer (17.18%, 61/355) come in second and third, respectively. While in Chinese cohort, the top three ARID1A mutation frequency tumor types were endometrial cancer (39.29%, 88/224), gastric carcinoma (17.80%, 318/1787) and urothelial carcinoma (17.18%, 83/483), respectively. ARID1A gene mutation was also associated with higher TMB in the Chinese pan-cancer cohort (P &lt; 0.0001). The highest medium TMB level of ARID1A mutation tumor type was Urothelial carcinoma with 18.63 Muts/Mb (n = 65). In addition, the TMB level and prognositic analysis were performed on patients in two independent NSCLC cohorts with ICIs, TMB level of ARID1A mutant group was higher than wild-type group with significant difference (P &lt; 0.0001). The overall survival (OS) of ARID1A mutation group were significantly shorter than wild-type group (OS, median, 6.80 vs 10.28 months; HR, 1.47; P = 0.0474), and a decreasing trend on progression-free survival (PFS) without significant difference (median, 1.46 vs 2.99 months; HR, 1.27; P = 0.1584). Conclusions: The results indicated that ARID1A gene mutation was associated with a higher TMB level in Chinese pan-cancer patients, and patients harboring these genes mutations might easily benefit from ICIs.

Evaluation of spatiotemporal heterogeneity of tumor mutational burden (TMB) in gastroesophageal adenocarcinoma (GEA) at baseline diagnosis and after chemotherapy.

4546 Background: Tumor mutational burden (TMB) may be a predictive marker for response to anti-PD1/PDL1 agents (IO). Molecular heterogeneity of various biomarkers for GEA has been established. To characterize heterogeneity of TMB and its clinical relevance, we compared TMB in primary (10) &amp; metastatic (met) tumors at baseline newly diagnosed stage IV advanced GEA (aGEA), and before &amp; after chemotherapy treatment (tx) for stage II–IV GEA. We assessed the prognostic relevance of TMB in aGEA. Methods: We retrospectively reviewed a cohort of 127 patients (pts) diagnosed with GEA in 2012–2019, for a total of 280 tumor samples with TMB data. TMB level was defined as low (≤5/Mb), intermediate (int) ( &gt; 5/Mb, ≤15/Mb), or high (hi) (≥15/Mb), determined by Foundation One. Analysis was performed by Fisher’s exact test for PDL1/TMB, McNemar’s test for paired TMB, and univariate Cox proportional-hazards model for overall survival (OS). Results: Of 280 tumors, 50% (140/280) had low TMB, 45% (125/280) int TMB, &amp; 5% (15/280) hi TMB. TMB ranged 0–58.6/Mb (median 5.3/Mb). Of tumors with hi TMB, 53% (8/15) were MSI-Hi, while of MSI-Hi tumors, 100% (8/8) were TMB hi. TMB level did not correlate with PDL1 status ( p= 0.83). Concordance between TMB levels of paired baseline 10 and baseline met tumors was 66% (29/44) (Table). TMB level was lower in the met than in the 10 in 23% (10/44) of cases, and higher in the met in 11% (5/44). Of 4 TMB hi baseline 10 tumors, 2 were not TMB hi in the met; of 40 TMB low/int baseline 10 tumors, 0 were TMB hi in the met ( p= 0.16). Post-tx tumors exhibited 71% (42/59) concordance of TMB levels compared to pre-tx 10 tumors. Of 2 TMB hi baseline tumors, 1 was not TMB hi in the post-tx tumor; of 57 TMB low/int baseline tumors, 0 were TMB hi in the post-tx tumor ( p= 0.32). In pts with aGEA at diagnosis, OS did not significantly differ depending on baseline 10 tumor TMB level (median OS of 21.4 [95% CI 15.4–27.9] months for TMB low, 14.6 [10.9–23.5] months for TMB int, and 9.6 [3.9–NA] for TMB hi; p= 0.3), nor depending on baseline met TMB level. Conclusions: Notable baseline spatial discordance of TMB was observed, particularly TMB hi 10 to low/int met. Discordance was also observed before &amp; after tx, without significant increase towards TMB hi temporally. Spatiotemporal heterogeneity may impact the role of TMB as a predictive biomarker &amp; warrants further study. [Table: see text]

The genomic landscape of immune-related genes and its correlations with TMB and PD-L1 expression in Chinese NSCLC.

e21056 Background: Understanding the genomic landscape of immune-related genes (IRGs) of patient with Non-Small Cell Lung Carcinoma (NSCLC) may provide critical insight and facilitate development of novel strategies for cancer therapy. Methods: Targeted next generation sequencing (NGS) was performed on tissue obtained from patients with NSCLC using a 381-gene panel. The IRGs were mainly based on a published article that summarized the genes related to activated T cells, immune cytolytic activity and IFN-γ release. Genomic alterations including single nucleotide variation (SNV), insertions/deletions, copy number variations (CNV) and gene fusions were assessed. Mutational profiles of IRGs and its correlations with Tumor mutation burden (TMB) and PD-L1 expression in NSCLC patients were analyzed from the 3D Medicine database. Results: Genomic data of 2,510 patients with NSCLC were analyzed via tissue detection [median age 62 years, (IQR: 54 - 69 years)]. 4.26% (107 / 2,510) of the patients had at least one characterized altered IRGs. The most frequently mutated IRGs identified for all genomic alternations occurred in PDCD1LG2 (40%), FAS (27%), LCK (23%), CD274 (21%), CD74 (3%), and missense mutations were the mainly mutant types of PDCD1LG2 and LCK, and copy number gains were showed in PDCD1LG2 and CD274. TMB levels in NSCLC patients with IRGs mutations were higher than in wild-type patients (mean TMB: 14.17 mut/MB vs 9.27 mut/MB, P &lt; 0.001). 65.42% (70/107, 9.35% for PD-L1 &gt; = 1% and &lt; 5%, 28.97% for PD-L1 &gt; = 5% and &lt; 50%, 27.10% for PD-L1 &gt; = 50%) of patients harboring PD-L1-positive ( &gt; = 1%), and the proportion of patients with mutations in IRGs does not exceed 10% in each subgroup. Results of Pearson Chi-square test showed that the PD-L1 expression had different between two groups ( Χ2 = 13.3774, P = 0.0038), patients harboring IRGs mutation had a significantly higher level of PD-L1 than those wide-type patients (P = 0.001). Conclusions: Genomic landscape of IRGs significantly correlated with that of PD-L1 expression. Further analyses using more precisely IRGs and different stage NSCLC samples are requisite to support a role for IRGs as an target to immune therapy of cancers.

Association between homologous recombination deficiency and tumor mutational burden in lung cancer.

e21043 Background: Homologous recombination (HR) is an important repair method for DNA double-strand damage. HR is involved with complex signaling pathways and multiple steps, including BRCA1/2. Homologous Recombination Deficiency (HRD) can be caused by the loss of function of BRCA1/2 proteins due to gene mutation. Tumor mutational burden (TMB) was indicated to involved with HRD, which is critical to the guidance of immunotherapy. Methods: Patients available with tumor specimen genomic testing for lung cancer were enrolled in this study. The sequencing library was captured using a 605-gene panel. Homologous recombination (HR)-related gene list included 102 genes. Genomic alterations of HR-related genes were assessed by next-generation sequencing assay, including nonsense, nonstop, frameshift, splice site, damaging missense mutations in somatic variants and pathogenic germline variants. Then, TMB was calculated by dividing the total number of mutations counted by the size of the coding region. Results: A total of 741 patients was enrolled and 182 of them (25.6%) had at least one genomic alteration of the HR genes. The top mutant HR genes included ATM (4.1%), DNMT3A (2.8%), ATR (1.8%), CHEK2 (1.5%), BRCA2 (1.5%), ABL1 (1.4%), FANCA (1.2%), RIF1 (1.1%), FANCI (1.1%), RAD50 (0.9%), BRCA1 (0.9%), and MRE11A (0.9%). The most common mutational type was missense mutation (69.2%), followed by frameshift (15.6%), and nonsense mutation (15.2%). The median age was 62 in the HRD group (age &lt; 45:5.4%, 45-65:59.6%, &gt; 65:34.9%) compared to 60 in the rest of the patients (age &lt; 45:11.3%, 45-65:61.3%, &gt; 65:27.4%). There were overall 57.5% males and 42.5% females in terms of gender. HRD group had a significantly higher rate than Non-HRD group in males (72.1% vs 52.8%, p &lt; 0.01). The occurrence of HRD mutations was significantly correlated with a high level of TMB (p &lt; 0.01). The median TMB of HRD group (7.9 muts/Mb, 95%CI:7.1-9.2 muts/Mb) was significantly higher than that of Non-HRD group (3.8 muts/Mb, 95%CI:3.3-4.2 muts/Mb). In addition, the upper quantile value (7.5 muts/Mb) was used to identify patients with high TMB. HRD group had a significantly higher rate of TMB-high patients than Non-HRD group (51.1% vs 17.0%, p &lt; 0.01). Conclusions: Our study demonstrated genomic alterations of the HR genes in about one fourth’s lung cancer. Besides, there was a significant positive correlation between HRD mutations and TMB level. The significance of HRD genomic mutation on predicting the efficacy of immunotherapy deserves further study.

Hepatocellular carcinoma with ARID1A mutation is associated with higher TMB and poor survival.

e16667 Background: Hepatocellular carcinoma (HCC) is a heterogeneous disease that lacks molecular predictors of response to available treatments. AT-rich interactive domain 1A (ARID1A), a SWI/SNF chromatin remodeling gene, is commonly mutated in human cancers and hypothesized to be a tumor suppressor gene. In recent years, immune checkpoint blockade immunotherapies (ICIs) are promising therapies for multiple cancers including HCC, and tumor mutation burden (TMB) was an important biomarker to predict the efficacy of ICIs. Besides, previous research reported that ARID1A deficiency was associated with mismatch repair (MMR) in cancer, and may cooperated with ICIs. Methods: Whole exome sequencing data of 10336 pan-cancer patients including 353 HCC patients was obtained from the Cancer Genome Altas (TCGA). Next generation sequencing (NGS) data of 15849 pan-cancer patients including 1926 HCC patients from Chinese clinical dataset were analyzed to explore the correlation between ARID1A gene mutation and TMB. TMB was defined as total number of somatic non-synonymous mutations in coding region. Prognostic data of 136 HCC patients were obtained from the MSK-IMPACT Clinical Sequencing Cohort (MSKCC). Results: In total, 8.62% (891/10336) of pan-cancer patients in TCGA harboring ARID1A mutation and 8.47% (3188/37628) in Chinese cohort, while mutant percentage of HCC was 9.35% (33/353) in TCGA, 9.03% (174/1926) in Chinese cohort, the alternation frequency of ARID1A in two cohorts was no significant difference (P = 0.3334). The medium TMB level of ARID1A mutant group was higher than wild-type group with significant difference both in Chinese pan-cancer and HCC (medium TMB level, P &lt; 0.0001). Results of 18 kinds of ARID1A mutated tumors patients in TCGA cohort indicated that ARID1A mutation was an independent risk factor in liver cancer affecting overall survival (OS, HR 2.43, 95% IC 1.07-5.54, P = 0.0286). Besides, Kaplan-Meier analysis was performed on HCC patients in MSKCC cohort, ARID1A statue resulted in significantly shorter OS (median, 12.2 vs 21.43 months; HR 2.5; P = 0.0092). Conclusions: The results indicated that ARID1A gene mutation was a poor prognostic factor in hepatocellular carcinoma, and these mutational states were associated with higher TMB level, which might be a potential positive biomarker of immunotherapy.

Acquired RB1 mutations in estrogen receptor-positive (ER+) clinically advanced and metastatic breast cancer (MBC).

1053 Background: Prior to use of cell cycle inhibitor drugs such as palbociclib, mutations in the retinoblastoma cell cycle inhibitory gene ( RB1) were extremely uncommon. To assess recent descriptions of RB1 mutations in MBC as treatment related, we compared primary untreated ER+ breast cancer samples from patients who developed MBC with ER+ MBC metastasis biopsies in post-primary treatment patients. Methods: Extracted DNA from 15 untreated primary breast tumors (PBx) and 36 treated metastatic MBC tumors (MBx) were sequenced by hybrid capture genomic profiling to evaluate all classes of genomic alterations (GA). Tumor mutational burden (TMB) was determined on up to 1.1 Mbp of sequenced DNA. PD-L1 expression was determined by IHC (Dako 22C3). Results: The age distribution, GA per tumor, CDH1 and ERBB2 GA frequencies were similar in PBx and MBx. GA in ESR1 (P &lt; 0.0008), classically associated with acquired hormonal therapy resistance, and PIK3CA (p &lt; 0.11) were increased in the MBx vs the PBx samples. RB1 GA were more frequent in the MBx samples (11%) than the pre-treatment PBx samples (0%, P &lt; 0.31). GA not associated with targeted therapy programs were similar in both groups. In addition to low frequencies of ERBB2 amplification in the ER+ tumors, the most frequent potentially targetable GA identified in both PBx and MBx included GA in FGFR1 and NF1. BRCA1 GA were identified in the MBx group only. Biomarkers of potential immunotherapy benefit including MSI status, TMB levels and PD-L1 IHC staining were at similarly low levels in both groups. Additional cases are being identified to expand the cohort sizes for final presentation. Conclusions: In addition to the well-known acquisition of hormonal therapy resistance mutations in ESR1 and PIK3CA in ER+ MBC, this study also supports the acquisition of RB1 mutations when post-treatment metastasis biopsies are sequenced rather than pre-treatment primary tumors. This increase in RB1 mutation identification is associated with exposure of the tumors to anti-cell cycle drugs such as palbociclib in the treated but not the untreated cohort. [Table: see text]

Germline and somatic DNA damage repair gene mutations potentially predict the efficacy of relevant treatment in Chinese patients with pancreatic ductal adenocarcinoma.

e16732 Background: PDAC is a fatal disease with molecular heterogeneity, inducing differences in biological behaviour and therapeutic strategy. We conducted a study to reveal the mutation landscape of Chinese PDAC patients, and investigate the predictive role of germline and somatic DNA damage repair (DDR) status in precise treatment. Methods: 195 PDAC patients were enrolled from multiple medical centers of China between Jan 2016 to Nov 2019. Baseline clinical or genetic characteristics, and survival status were collected. NGS were performed on paraffin-embedded resected tissues or peripheral blood using a panel of 417 genes, including 50 DDR-related genes. Survival analysis was conducted using Kaplan-Meier, and Cox proportional hazard regression model. Results: The main driver genes were KRAS, TP53, CDKN2A, and SMAD4. Patients with KRAS mutation showed worse OS than those without (p = 0.048). DDR deficiency were identified in 15.38% of overall patients, mainly occurred in BRCA2 (4.62%), ATM (4.10%), RAD50 (1.54%) and MLH1 genes (1.03%). No significant improvement of OS existed between patients with or without DDR mutations (p = 0.88). Treatment with olaparib (adjusted HR, 0.2550; P = 0.0720) or platinum-based chemotherapy (adjusted HR, 0.1308; P = 0.0185) respectively decreased hazard of death in patients with DDR mutation. Besides BRCA gene, ATM mutant patients treated with olaparib harbored prolonged median OS than those without olaparib treatment (22.25 vs 15.2 month). Despite a little higher PD-L1 expression rate were seen in DDR mutant patients (29.17% vs 20.51%), no statistical correlation between tumor mutation burden level and DDR mutation was identified. And in patients treated with PD-1 blockade, 2 patients of DDR wild-type group both had SD, whereas of the remaining 5 patients with DDR deficiency, 1 was evaluated as PR, 3 as SD, and 1 as PD (ORR, 0 wt vs 20% mut). Conclusions: In this multi-center retrospective study, we deciphered the intra-tumoral genetic heterogeneity in Chinese PDAC population, which differs from western patients cohort to some extent. We found the potential role of germline and somatic DDR mutation status in predicting the response to olaparib and platinum-based chemotherapy, especially with BRCA or ATM mutation. However, DDR alteration was limited in prediction of hypermutational status and sensitivity to PD-1 blockade. Our study may provide a valuable evidence for clinical application of DDR mutation as a potential biomarker for precise treatment.

Predictive and prognostic impact of primary tumor-bearing lobe in nonsmall cell lung cancer patients treated with anti-PD-1 therapy.

Immunotherapy targeting programmed cell death-1 (PD-1) has become a standard pharmacological therapy. Although tumor mutation burden level was reported to depend on the tumor location in nonsmall cell lung cancer (NSCLC), predictive impact of the tumor location on the response to anti-PD-1 therapy is unknown. Two hundred and seventeen advanced or recurrent NSCLC patients treated with anti-PD-1 therapy at Kyushu University Hospital and National Hospital Organization Kyushu Cancer Center were analyzed. To minimize the bias arising from the patients' background, adjusted Kaplan-Meier survival curves and Cox proportional hazards regression analyses using inverse probability of treatment weights (IPTW) were performed. Of the 217 patients, 132, 27, and 58 had primary NSCLC in upper, middle, and lower lobes, respectively. Patients with NSCLC in upper lobe were significantly associated with younger age (P = .0070) and smoker (P = .0003). The epidermal growth factor receptor-wild type and tumor location in upper lobe were independent predictors of disease control (P = .0175 and P = .0425, respectively). The IPTW-adjusted Kaplan-Meier curves showed that patients with NSCLC in the upper lobes had significantly longer progression-free survival (PFS) and overall survival (OS) than those in middle/lower lobes (P = .0026 and P = .0015, respectively). On IPTW adjusted Cox analysis, NSCLC in the upper lobe was an independent predictor of PFS and OS (P = .0078 and P = .0034, respectively). Patients with primary NSCLC in the upper lobes may be good candidates for anti-PD-1 therapy. These findings should be validated prospectively.

High tumor mutation burden is associated with DNA damage repair gene mutation in breast carcinomas

BackgroundImmunotherapy has demonstrated encouraging clinical benefits in patients with advanced breast carcinomas and Programmed death ligand 1 (PD-L1) expression has been proposed as an immunotherapy biomarker. Challenges with current PD-L1 testing exist and tumor mutation burden (TMB) is emerging as a biomarker to predict clinical response to immunotherapy in melanoma and non-small cell lung cancer patients. However, TMB has not been well characterized in breast carcinomas.MethodsThe study cohort included 62 advanced breast cancer patients (13 primary and 49 metastatic). Genetic alterations and TMB were determined by FoundationOne CDx next generation sequencing (NGS) and the association with clinicopathologic features was analyzed.ResultsHigh TMB was observed in a relatively low frequency (3/62, 4.8%). TMB levels were positively associated tumor infiltrating lymphocytes and significantly higher TMB was observed in breast carcinomas with DNA damage repair gene mutation(s). There was no significant association between TMB levels and other analyzed clinicopathologic characteristics.ConclusionsOur data indicate the importance of DNA damage repair proteins in maintaining DNA integrity and immune reaction and breast carcinoma patients with DDR mutation may benefit from immunotherapy.

TGFB2 serves as a link between epithelial-mesenchymal transition and tumor mutation burden in gastric cancer

Immune checkpoint blockade (ICB) has been a major breakthrough in various cancers including gastric cancer (GC), yet the clinical outcomes remain poor. Currently, epithelial-mesenchymal transition (EMT) has been reported to be associated with tumor mutational burden (TMB), which can cause lack of response to ICB. However, the underlying mechanism remains unknown. Members of the transforming growth factorβ (TGFB) family are regarded as the main mediators of EMT, yet how TGFB2 drives EMT in GC is not fully understood. In this study, we found that overexpression of TGFB2 was correlated with poor prognosis in TGCA-STAD and four GEO GC datasets.Gene set enrichment analysis revealed that the EMT pathway was significantly enriched in the high TGFB2 expression group, whilst the TMB-related pathways including mismatch repair, base excision repair, and DNA replication were strongly enriched in the low expression group. Furthermore, EMT score analysis, WGCNA and functional analysis showed that TGFB2 was co-expressed with neurite-related pathways that might drive EMT. Also, CIBERSORT analysis revealed that tumor-infiltrating immune cells like T follicular helper cells might participate in the process of TGFB2 affecting TMB levels in GC. Moreover, in other various cancers, TGFB2 was also negatively correlated with TMB levels as well as ICB response. Overall, these results revealed that TGFB2 could play a vital role in linking EMT and TMB in GC, suggesting that TGFB2 may be a predictive therapeutic target for GC.

Association of genetic and immuno-characteristics with clinical outcomes in patients with RET-rearranged non-small cell lung cancer: a retrospective multicenter study

BackgroundRearranged during transfection (RET) has been proven to be a tumorigenic target in non-small cell lung cancers (NSCLCs). In RET-rearranged NSCLCs, molecular features and their impact on prognosis were not well illustrated, and the activity of mainstay therapeutics has not currently been well compared.MethodsPatients diagnosed with NSCLCs with RET rearrangements were analyzed for concomitant mutations, tumor mutation burden (TMB), PD-L1 expression, T cell receptor repertoire and clinical outcomes with chemotherapy, immune checkpoint inhibitors (ICIs), and multikinase inhibitors (MKIs).ResultsAmong 129 patients with RET-rearranged NSCLC who were analyzed, 41.1% (53/129) had co-occurring genetic alterations by next-generation sequencing, and concomitant TP53 mutation appeared most frequently (20/53, 37.7%). Patients with concurrent TP53 mutation (n = 15) had shorter overall survival than those without (n = 30; median, 18.4 months [95% CI, 8.6–39.1] vs 24.8 months [95% CI, 11.7–52.8]; P < 0.05). Patients with lower peripheral blood TCR diversity (n = 5) had superior overall survival compared with those with higher diversity (n = 6; median, 18.4 months [95% CI, 16.9–19.9] vs 4.8 months [95% CI, 4.5–5.3]; P = 0.035). An association with overall survival was not observed for PD-L1 expression nor for tumor mutation burden level. Median progression-free survival was not significantly different across chemotherapy, ICIs, and MKIs (median, 3.5 vs 2.5 vs 3.8 months). For patients treated with ICIs, the disease control rate was 60% (6/10) and the objective response rate was 20% (2/10).ConclusionsRET-rearranged lung cancers can be heterogeneous in terms of concomitant genetic alterations. Patients with concurrent TP53 mutation or high peripheral blood TCR repertoire diversity have relatively inferior overall survival in this series. Outcomes with traditional systemic therapies in general are suboptimal.

Tumor Mutation Burden and Depression in Lung Cancer: Association With Inflammation.

Patients with lung cancer with greater systemic inflammation have higher rates of depression. Tumor mutation burden (TMB) predicts immunotherapy response in patients with lung cancer and is associated with intratumoral inflammation, which may contribute to systemic inflammation and depression. This study evaluated whether higher TMB was associated with increased depression and systemic inflammation in patients with lung cancer. Patients with metastatic lung cancers were evaluated for depression severity using the Hospital Anxiety and Depression Scale. TMB was measured using the Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets. Inflammation was evaluated using C-reactive protein (CRP) level and neutrophil-to-lymphocyte ratio (NLR). A total of 96 patients with adequate TMB testing were evaluated. The average number of mutations (TMB) was 10.8 (SD, 10.9). A total of 19% of patients endorsed clinically significant depression symptoms. TMB was significantly correlated with depression severity (r = 0.34; P=.001) and NLR (r = 0.37; P=.002) but not CRP level (r = 0.19; P=.07). TMB was also higher in patients receiving chemotherapy (mean, 12.0) and immunotherapy (mean, 14.4) versus targeted therapy (mean, 4.8). A multivariate model found that TMB (β = 0.30; P=.01) and CRP level (β = 0.31; P=.01) were independently associated with depression; there was no significant interaction effect of TMB × CRP and depression. A similar multivariate model showed no independent effect for NLR and depression (β = 0.16; P=.17) after accounting for TMB. These data provide evidence for biologic depression risk in patients with lung cancer who have high levels of TMB. The underlying mechanism of the association is not clearly related to inflammation but warrants further analysis to broadly elucidate the mechanism of biologically derived depression in cancer.

NF2 mutation-driven renal cell carcinomas (RCC): A comprehensive genomic profiling (CGP) study.

726 Background: NF2 genomic alterations (GA) have been associated with aggressive behavior in RCC. Methods: FFPE tissues from 1,386 clear cell (ccRCC), 307 papillary (pRCC), 72 chromophobe (chRCC), 145 sarcomatoid (sRCC), 54 collecting duct (cdRCC),37 medullary (medRCC) and 134 unclassified (nosRCC) underwent hybrid-capture based CGP to evaluate all classes of genomic alterations (GA). Tumor mutational burden (TMB) was determined on up to 1.1 Mbp of sequenced DNA and MSI was determined on 114 loci. PD-L1 expression was determined by IHC (Dako 22C3). Results: 140 (7%) RCC featured NF2 GA which were predominantly short variant (SV) mutations. Gender and age were similar with male preponderance in all histologic subtypes. NF2 GA frequency was highest in cdRCC (20%) and sRCC (19%) and lowest in ccRCC (3%). The medRCC at 5% NF2 GA and chRCC at 0% NF2 GA were not further evaluated. VHL and PBRM1 GA were significantly more frequent in NF2 altered ccRCC than all other RCC (P &lt; 0.001). Other mTOR pathway GA were uncommon. Potentially targetable kinase GA in NF2-mutated RCC included BRAF (2% of ccRCC), EGFR (3% of pRCC), ERBB3 (4% of sRCC) and PIK3CA (9% of cdRCC). No NF2 mutated RCC featured MSI -high status and both TMB and PD-L1 expression levels were extremely low in all subsets with exception of high PD-L1 staining in sRCC tumors. Conclusions: cdRCC, sRCC, pRCC and nosRCC are enriched in NF2 GA. Low PBRM1 GA, TMB and MSI- high predict resistance to immunotherapy in NF2 mutated RCC although the high PD-L1 expression in sRCC is noteworthy.[Table: see text]

Comprehensive genomic profiling (CGP) of histologic subtypes of urethral carcinomas (UrthCa).

426 Background: UrthCa is an uncommon GU malignancy that can progress to advanced metastatic disease. Methods: 126 metastatic UrthCa underwent hybrid-capture based CGP to evaluate all classes of genomic alterations (GA). Tumor mutational burden (TMB) was determined on up to 1.1 Mbp of sequenced DNA and microsatellite instability (MSI) was determined on 114 loci. PD-L1 expression was determined by IHC (Dako 22C3). Results: 49 (39%) urothelial (UrthUC), 31 (24%) squamous (UrthSCC), 34 (27%) adenocarcinomas (UrthAC) and 12 (9%) clear cell (UrthCC) were evaluated. UrthUC and UrthSCC were more common in men; UrthAC and UrthCC more common in women. Ages were similar in all groups. GA in PIK3CA were the most frequent potentially targetable GA; mTOR pathway GA in PTEN also identified. GA in other potentially targetable genes were also identified including ERBB2 (6% in UrthUC, 3% in UrthSCC and 12% in UrthAC), FGFR1-3 (3% in UrthSCC), BRAF (3% in UrthAC), PTCH1 (8% in UrthCC) and MET (8% in UrthCC). Possibly reflecting their higher GA/tumor status, potential for immunotherapy benefit associated with higher TMB and PD-L1 staining levels were seen in UrthUC and UrthSCC compared to UrthAC and UrthCC. MSI high status was absent throughout. Conclusions: CGP reveals GA that may be predictive of both targeted and immunotherapy benefit in patients with advanced UrthCa and that could potentially be used in future adjuvant, neoadjuvant and metastatic disease trials.[Table: see text]

Comprehensive genomic profiling (CGP) in post-systemic treatment (Post) metastatic sites (MET) and pretreatment (Pre) primary tumors (PT) of metastatic prostate cancer (mPC).

175 Background: CGP was performed on Pre PT and Post MET including bone (BO), liver (LIV), lung (LU), brain (BN), lymph node (LN) and soft tissue (ST) mPC. Methods: 1,294 mPC underwent hybrid-capture based CGP. Tumor mutational burden (TMB) was determined on up to 1.1 Mbp of sequenced DNA and microsatellite instability (MSI) was determined on 114 loci. PD-L1 expression was determined by IHC (Dako 22C3). Results: GA in AR were lowest in the Pre PT (2%) and highest in Post LU (24%) and LIV (50%). GA/tumor was significantly higher in BN (8.0) compared to PT (3.8). BR MET also featured higher PTEN GA than PT. BRCA2 GA varied from 0% in BR to 7-9% in PT, BO, LU, LN and ST to a high of 15% in LI MET. ATM GA were significantly higher in LU MET and RAD21 GA highest in LN MET. Potential predictors of IO drug response included high CDK12 GA in LU MET, MSI high status at 29% in BR MET associated with higher TMB levels, but virtual absence of high PD-L1 expression. ERBB2 GA appeared to be increased in the MET group compared with PT but BRAF GA were not. RB1 GA were significantly increased in LIV MET cases. Conclusions: CGP of mPC PT and MET demonstrates significant differences likely linked to exposure to systemic therapies. These findings suggest that, in the future, liquid biopsies may have advantages over individual MET site biopsies in their ability to capture the entire range of therapeutic opportunities for patients with advanced mPC.[Table: see text]

Tumor microenvironment (TME), longitudinal biomarker changes, and clinical outcome in patients (pts) with advanced BRAF V600–mutant melanoma treated with first-line spartalizumab (S) + dabrafenib (D) + trametinib (T).

39 Background: Although pts with both low tumor mutation burden (TMB) and T-cell–inflamed gene expression profiles (TI-GEPs) usually have poor outcomes with anti–PD-1 therapy, an analysis in the adjuvant melanoma setting suggested that these pts benefited from adjuvant D+T therapy. Here we analyze TMB/TI-GEPs and other biomarkers in pts receiving a combination of anti–PD-1 and D+T therapy. Methods: The phase 3 COMBI-i study (NCT02967692) is evaluating S in combination with D+T in previously untreated pts with BRAF V600–mutant unresectable/metastatic melanoma. In the safety run-in (part [p] 1) and biomarker (p2) cohorts, blood/tissue samples were collected at baseline (BL), after 2-3 and 8-12 wk of treatment, and at disease progression. TMB/circulating tumor DNA (ctDNA) and TI-GEPs were examined by targeted DNA-seq and RNA-seq, respectively. Results: At data cutoff, 6 of 22 pts with DNA- and RNA-seq data available had a PFS event. At BL, these pts had low TMB, low TI-GEPs (4 of 6), or high levels of immunosuppressive TME signatures (eg, fibroblast, M2 macrophages) vs pts without a PFS event. Elevated BL ctDNA was significantly associated with PFS events ( P&lt; .001). Pts with a complete response (CR) on S+D+T had significantly lower levels of BL immunosuppressive TME signatures (eg, M2 macrophages; P&lt; .01) than pts without a CR. We observed a consistent increase in TI-GEPs and decrease in MAPK pathway activity score (MPAS) from BL to biopsy at 2-3 wk in all pts regardless of subsequent progression. Pts with a PFS event and available longitudinal biomarker data were characterized by a subsequent decrease in TI-GEPs and an increase in MPAS per the 8- to 12- wk biopsy sample. Conclusions: These results suggest that S+D+T had an early impact on tumor cells and the TME, potentially promoting antitumor activity. The majority of PFS events occurred in the TMB-low/TI-GEP-low subgroup. An immunosuppressive TME might preclude early CRs. The predictive implications of coupling TMB/GEP subgroups with other TME marker subgroups need further validation. The randomized placebo-controlled p3 of COMBI-i is ongoing. Clinical trial information: NCT02967692.

Multi-omics analysis of tumor mutational burden combined with prognostic assessment in epithelial ovarian cancer based on TCGA database.

Background: Tumor mutation burden (TMB) is considered as a novel biomarker of response to immunotherapy and correlated with survival outcomes in various malignancies. Here, TMB-related genes (TRGs) expression signatures were constructed to investigate the association between TMB and prognosis in epithelial ovarian cancer (EOC), and the potential mechanism in immunoregulation was also explored.Methods: Based on somatic mutation data of 436 EOC samples from The Cancer Genome Atlas database, we examined the relationship between TMB level and overall survival (OS), as well as disease-free survival (DFS). Next, the TRGs signatures were constructed and validated. Differential abundance of immune cell infiltration, expression levels of immunomodulators and functional enrichment in high- and low-risk groups were also analyzed.Results: Higher TMB level revealed better OS and DFS, and correlated with earlier clinical stages in EOCs (P = 2.796e-04). The OS-related prognostic model constructed based on seven TRGs (B3GALT1, LIN7B, ANGPT2, D2HGDH, TAF13, PFDN4 and DNAJC19) significantly stratified EOC patients into high- and low-risk groups (P < 0.001). The AUC values of the seven-gene prognostic signature at 1 year, 3 years, and 5 years were 0.703, 0.758 and 0.777. While the DFS-related prognostic model was constructed based on the 4 TRGs (LPIN3, PXYLP1, IGSF23 and B3GALT1), with AUCs of 0.617, 0.756, and 0.731, respectively. Functional analysis indicated that immune‐related pathways were enriched in low‐risk groups. When considering the infiltration patterns of immune cells, we found higher proportions of follicular helper T (Tfh) cell and M1 macrophage, while lower infiltration of M0 macrophage in low-risk groups (P < 0.05). Accordingly, TMB levels of low-risk patients were significantly higher both in OS and DFS model (P < 0.01).Conclusions: Our TRGs-based models are reliable predictive tools for OS and DFS. High TMB may confer with an immunogenic microenvironment and predict favorable outcomes in EOCs.