An interpretable deep learning model for identifying the morphological characteristics of dMMR/MSI-H gastric cancer.

Accurate tumor diagnosis by pathologists relies on identifying specific morphological characteristics. However, summarizing these unique morphological features in tumor classifications can be challenging. Although deep learning models have been extensively studied for tumor classification, their indirect and subjective interpretation obstructs pathologists from comprehending the model and discerning the morphological features accountable for classifications. In this study, we introduce a new approach utilizing Style Generative Adversarial Networks, which enables a direct interpretation of deep learning models to detect significant morphological characteristics within datasets representing patients with deficient mismatch repair/microsatellite instability-high gastric cancer. Our approach effectively identifies distinct morphological features crucial for tumor classification, offering valuable insights for pathologists to enhance diagnostic accuracy and foster professional growth.

Primary tumor resection with or without metastasectomy for left- and right-sided stage IV colorectal cancer: an instrumental variable analysis.

BACKGROUND: Whether primary tumor location (PTL) is predictive of survival benefits following primary tumor resection plus metastasectomy (PMTR) and primary tumor resection (PTR) alone in stage IV colorectal cancer patients is not known. We sought to address this issue by employing instrumental variable analysis to evaluate the efficacy of PMTR and PTR with stratification for primary tumor location in stage IV colorectal cancer patients. PATIENTS AND METHODS: Stage IV colorectal cancer patients diagnosed between January 1, 2005 and December 31, 2015 were identified from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute. To account for both measured and unmeasured confounders, the efficacy of PMTR and PTR in the left- and right-sided subgroups was evaluated using instrumental variable analysis, with the health service area as the instrument variable. Overall survival (OS) was the primary outcome of interest. RESULTS: A total of 50,333 eligible patients were analyzed (left-sided, n = 29,402 and right-sided, n = 20,931). OS was significantly better with PMTR than with other treatments (PTR, metastasectomy only, or no surgery) in patients with left-sided tumors (hazard ratio [HR] = 0.37 [95% CI 0.24-0.58], P < 0.001), but not in patients with right-sided tumors (HR = 0.98 [95% CI 0.65-1.47], P = 0.910; interaction test P < 0.001). OS was comparable in patients treated with PTR and those treated with no surgery in both the left-sided (HR = 1.11 [95% CI 0.68-1.81], P = 0.690) and right-sided (HR = 0.85 [95% CI 0.50-1.43], P = 0.530; interaction test P = 0.466) subgroups. CONCLUSIONS: PMTR appears to only benefit patients with left-sided stage IV colorectal cancer but not those with right-sided tumors. PTR does not improve OS, regardless of primary tumor location. When selecting patients for PMTR, primary tumor location should be considered. Overuse of PTR should be avoided.

Circulating tumor DNA methylation marker MYO1-G for diagnosis and monitoring of colorectal cancer.

BACKGROUND: Circulating tumor DNA (ctDNA) is a promising diagnostic and prognostic marker for many cancers and has been actively investigated in recent years. Previous studies have already demonstrated the potential use of ctDNA methylation markers in the diagnosis and prognostication of colorectal cancer (CRC). This retrospective study validated the value of methylation biomarker MYO1-G (cg10673833) in CRC diagnosis and disease monitoring using digital droplet PCR (ddPCR), a biomarker selected from our previous study due to its highest diagnostic efficiency. METHODS: Blood samples of CRC and control samples from tumor-free individuals at two institutions were collected to quantify the methylation ratio using ddPCR. Area under curve (AUC) was calculated after constructing receiver operating characteristic curve (ROC) for CRC diagnosis. Sensitivity and specificity were estimated and comparisons of methylation ratio in different groups were performed. RESULTS: We collected 673 blood samples from 272 patients diagnosed with stage I-IV CRC and 402 normal control samples. The methylation biomarker discriminated patients with CRC from normal controls with high accuracy (area under curve [AUC] = 0.94) and yielded a sensitivity of 84.3% and specificity of 94.5%. Besides, methylation ratio of MYO1-G was associated with tumor burden and treatment response. The methylation ratio was significantly lower in patients after their radical operation than when compared with those before surgeries (P < 0.001). Methylation ratio was significantly higher in patients with disease progression than those with stable disease (P = 0.002) and those with complete response or partial response (P = 0.009). CONCLUSIONS: Together, our study indicated that this methylation marker can serve as a potential biomarker for diagnosing and monitoring CRC.

Exploration of modified progression-free survival as a novel surrogate endpoint for overall survival in immuno-oncology trials.

BACKGROUND: Progression-free survival (PFS) exhibits suboptimal performance as the surrogate endpoint for overall survival (OS) in trials studying immune checkpoint inhibitors (ICIs). Here we propose a novel surrogate endpoint, modified PFS (mPFS), which omits the events of disease progression (but not deaths) within 3 months after randomization. METHODS: PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials were searched for randomized trials studying ICIs in advanced solid tumors with available PFS and OS data up to May 2020. Individual patient-level data (IPD) for PFS and OS were reconstructed for eligible trials. A simulation-based algorithm was used to match the reconstructed, disconnected PFS and OS IPD, and 1000 independent simulated datasets of matched PFS-OS IPD were generated for each trial. mPFS durations and statuses were then measured for each of the matched PFS-OS IPD. Trial-level correlation between Cox HRs for PFS or mPFS and HRs for OS was assessed using Pearson correlation coefficient (rp) weighted by trial size; patient-level correlation between PFS or mPFS and OS was assessed using Spearman's rank correlation coefficient (rs). Findings were further validated using the original IPD from two randomized ICI trials. RESULTS: Fifty-seven ICI trials totaling 29,429 participants were included. PFS HR showed moderate correlation with OS HR (rp=0.60), and PFS was moderately correlated with OS at the patient level (median rs=0.66; range, 0.65-0.68 among the 1000 simulations). In contrast, mPFS HR achieved stronger correlation with OS HR (median rp=0.81; range, 0.77-0.84), and mPFS was more strongly correlated with OS at the patient level (median rs=0.79; range, 0.78-0.80). The superiority of mPFS over PFS remained consistent in subgroup analyses by cancer type, therapeutic regimen, and treatment setting. In both trials with the original IPD where experimental treatment significantly improved OS, mPFS successfully captured such clinical benefits whereas PFS did not. CONCLUSIONS: mPFS outperformed PFS as the surrogate endpoint for OS in ICI trials. mPFS is worthy of further investigation as a secondary endpoint in future ICI trials.

EPHA7 mutation as a predictive biomarker for immune checkpoint inhibitors in multiple cancers.

BACKGROUND: A critical and challenging process in immunotherapy is to identify cancer patients who could benefit from immune checkpoint inhibitors (ICIs). Exploration of predictive biomarkers could help to maximize the clinical benefits. Eph receptors have been shown to play essential roles in tumor immunity. However, the association between EPH gene mutation and ICI response is lacking. METHODS: Clinical data and whole-exome sequencing (WES) data from published studies were collected and consolidated as a discovery cohort to analyze the association between EPH gene mutation and efficacy of ICI therapy. Another independent cohort from Memorial Sloan Kettering Cancer Center (MSKCC) was adopted to validate our findings. The Cancer Genome Atlas (TCGA) cohort was used to perform anti-tumor immunity and pathway enrichment analysis. RESULTS: Among fourteen EPH genes, EPHA7-mutant (EPHA7-MUT) was enriched in patients responding to ICI therapy (FDR adjusted P < 0.05). In the discovery cohort (n = 386), significant differences were detected between EPHA7-MUT and EPHA7-wildtype (EPHA7-WT) patients regarding objective response rate (ORR, 52.6% vs 29.1%, FDR adjusted P = 0.0357) and durable clinical benefit (DCB, 70.3% vs 42.7%, FDR adjusted P = 0.0200). In the validation cohort (n = 1144), significant overall survival advantage was observed in EPHA7-MUT patients (HR = 0.62 [95% confidence interval, 0.39 to 0.97], multivariable adjusted P = 0.0367), which was independent of tumor mutational burden (TMB) and copy number alteration (CNA). Notably, EPHA7-MUT patients without ICI therapy had significantly worse overall survival in TCGA cohort (HR = 1.33 [95% confidence interval, 1.06 to 1.67], multivariable adjusted P = 0.0139). Further gene set enrichment analysis revealed enhanced anti-tumor immunity in EPHA7-MUT tumor. CONCLUSIONS: EPHA7-MUT successfully predicted better clinical outcomes in ICI-treated patients across multiple cancer types, indicating that EPHA7-MUT could serve as a potential predictive biomarker for immune checkpoint inhibitors.