Kirsch, Richard

Abstract Deep learning may detect biologically important signals embedded in tumor morphologic features that confer distinct prognoses. Tumor morphologic features were quantified to enhance patient risk stratification within DNA mismatch repair (MMR) groups using deep learning. Using a quantitative segmentation algorithm (QuantCRC) that identifies 15 distinct morphologic features, we analyzed 402 resected stage III colon carcinomas [191 deficient (d)-MMR; 189 proficient (p)-MMR] from participants in a phase III trial of FOLFOX-based adjuvant chemotherapy. Results were validated in an independent cohort (176 d-MMR; 1,094 p-MMR). Association of morphologic features with clinicopathologic variables, MMR, KRAS, BRAFV600E, and time-to-recurrence (TTR) was determined. Multivariable Cox proportional hazards models were developed to predict TTR. Tumor morphologic features differed significantly by MMR status. Cancers with p-MMR had more immature desmoplastic stroma. Tumors with d-MMR had increased inflammatory stroma, epithelial tumor-infiltrating lymphocytes (TIL), high-grade histology, mucin, and signet ring cells. Stromal subtype did not differ by BRAFV600E or KRAS status. In p-MMR tumors, multivariable analysis identified tumor-stroma ratio (TSR) as the strongest feature associated with TTR [HRadj 2.02; 95% confidence interval (CI), 1.14–3.57; P = 0.018; 3-year recurrence: 40.2% vs. 20.4%; Q1 vs. Q2–4]. Among d-MMR tumors, extent of inflammatory stroma (continuous HRadj 0.98; 95% CI, 0.96–0.99; P = 0.028; 3-year recurrence: 13.3% vs. 33.4%, Q4 vs. Q1) and N stage were the most robust prognostically. Association of TSR with TTR was independently validated. In conclusion, QuantCRC can quantify morphologic differences within MMR groups in routine tumor sections to determine their relative contributions to patient prognosis, and may elucidate relevant pathophysiologic mechanisms driving prognosis. Significance: A deep learning algorithm can quantify tumor morphologic features that may reflect underlying mechanisms driving prognosis within MMR groups. TSR was the most robust morphologic feature associated with TTR in p-MMR colon cancers. Extent of inflammatory stroma and N stage were the strongest prognostic features in d-MMR tumors. TIL density was not independently prognostic in either MMR group.

Abstract Purpose: There is a need to improve current risk stratification of stage II colorectal cancer to better inform risk of recurrence and guide adjuvant chemotherapy. We sought to examine whether integration of QuantCRC, a digital pathology biomarker utilizing hematoxylin and eosin–stained slides, provides improved risk stratification over current American Society of Clinical Oncology (ASCO) guidelines. Experimental Design: ASCO and QuantCRC-integrated schemes were applied to a cohort of 398 mismatch-repair proficient (MMRP) stage II colorectal cancers from three large academic medical centers. The ASCO stage II scheme was taken from recent guidelines. The QuantCRC-integrated scheme utilized pT3 versus pT4 and a QuantCRC-derived risk classification. Evaluation of recurrence-free survival (RFS) according to these risk schemes was compared using the log-rank test and HR. Results: Integration of QuantCRC provides improved risk stratification compared with the ASCO scheme for stage II MMRP colorectal cancers. The QuantCRC-integrated scheme placed more stage II tumors in the low-risk group compared with the ASCO scheme (62.5% vs. 42.2%) without compromising excellent 3-year RFS. The QuantCRC-integrated scheme provided larger HR for both intermediate-risk (2.27; 95% CI, 1.32–3.91; P = 0.003) and high-risk (3.27; 95% CI, 1.42–7.55; P = 0.006) groups compared with ASCO intermediate-risk (1.58; 95% CI, 0.87–2.87; P = 0.1) and high-risk (2.24; 95% CI, 1.09–4.62; P = 0.03) groups. The QuantCRC-integrated risk groups remained prognostic in the subgroup of patients that did not receive any adjuvant chemotherapy. Conclusions: Incorporation of QuantCRC into risk stratification provides a powerful predictor of RFS that has potential to guide subsequent treatment and surveillance for stage II MMRP colorectal cancers.

162 Background: There is a need to improve current risk stratification of stage II and III colorectal cancer (CRC) to better inform risk of recurrence and guide adjuvant chemotherapy. The purpose of this study is to examine whether integration of QuantCRC, an AI-based digital pathology biomarker utilizing hematoxylin and eosin-stained slides, provides improved risk stratification over current American Society of Clinical Oncology (ASCO) guidelines. Methods: ASCO and QuantCRC-integrated risk schemes were applied to an observational cohort of 1,068 stage II and III CRCs. The stage II integrated scheme utilizes pT3 vs. pT4 and QuantCRC-derived risk groups. The stage III integrated scheme utilizes pT1-3 vs. pT4, pN1 vs. pN2, and QuantCRC-derived risk groups. Performance metrics included log-rank test, hazard ratios and Somers’ Dxy rank correlation. Results: Integration of QuantCRC provides improved risk stratification compared to the ASCO scheme for stage II and III CRC. The QuantCRC-integrated scheme placed more stage II tumors in the low-risk group compared to the ASCO scheme (69.3% vs. 60.4%) without decreased 3-year RFS. The QuantCRC-integrated scheme provided larger hazard ratios (HR) for both intermediate-risk (3.04, 95%CI 1.81-5.10, P=2.8x10-5) and high-risk (4.62, 95%CI 2.02-10.61, P=0.0003) groups compared to ASCO intermediate-risk (2.09, 95%CI 1.21-3.63, P=0.008) and high-risk (3.08, 95%CI 1.57-6.01, P=0.001) groups. The QuantCRC-integrated scheme for stage III tumors identified a small group of 80/518 (15.4%) CRCs at very high risk of recurrence with HR of 4.08 (95%CI 2.68-6.23, P=6.5x10-11) compared to a HR of 2.42 (95%CI 1.72-3.38, P=3.1x10-7) for 228/518 (44.0%) high-risk CRCs in the ASCO scheme. QuantCRC-integrated risk groups remained prognostic in stage III CRCs when stratified by presence or absence of any adjuvant chemotherapy. No difference in RFS were seen in QuantCRC-integrated low-risk and intermediate-risk stage III CRCs stratified by 3 vs. 6-months of oxaliplatin-based adjuvant chemotherapy suggesting that these two groups can be treated with 3-months of adjuvant therapy. Conclusions: Incorporation of QuantCRC into risk stratification provides a powerful predictor of RFS that has potential to guide subsequent treatment and surveillance.

Immune checkpoint inhibitor-associated colitis (ICIC) affects approximately 15% of cancer patients treated with immunotherapy. Although histological evaluation is potentially valuable for both the diagnosis of ICIC and evaluation of disease activity, use in clinical practice is heterogeneous. We aimed to develop expert recommendations to standardize histological assessment of disease activity in patients with ICIC. Using the modified Research and Development/University of California Los Angeles (RAND/UCLA) appropriateness methodology, an international panel of 11 pathologists rated the appropriateness of 99 statements on a 9-point Likert scale during two rounds of anonymous voting. Results were discussed between rounds using moderated videoconferences. There are currently no disease-specific instruments for assessing histological features of ICIC. The panel considered that colonoscopy with at least three biopsies per segment from a total of at least five segments, including both endoscopically normal and inflamed areas, was appropriate for tissue acquisition. They agreed that biopsies should be oriented such that the long axis of the colonic crypts is visualized and should be stained with hematoxylin and eosin. Histological items that the panel voted were appropriate to evaluate in ICIC included the degree of structural/architectural change, chronic inflammatory infiltrate, lamina propria and intraepithelial neutrophils, crypt abscesses and destruction, erosions/ulcerations, apoptosis, surface intraepithelial lymphocytosis, and subepithelial collagen thickness. The appropriateness of routine immunohistochemistry was uncertain. These expert recommendations will help standardize assessment of histological activity in patients with ICIC. The panel also identified the development and validation of an ICIC-specific histological index as a research priority.

AimsTo develop and validate a deep learning algorithm to quantify a broad spectrum of histological features in colorectal carcinoma.Methods and resultsA deep learning algorithm was trained on haematoxylin and eosin‐stained slides from tissue microarrays of colorectal carcinomas (N = 230) to segment colorectal carcinoma digitised images into 13 regions and one object. The segmentation algorithm demonstrated moderate to almost perfect agreement with interpretations by gastrointestinal pathologists, and was applied to an independent test cohort of digitised whole slides of colorectal carcinoma (N = 136). The algorithm correctly classified mucinous and high‐grade tumours, and identified significant differences between mismatch repair‐proficient and mismatch repair‐deficient (MMRD) tumours with regard to mucin, inflammatory stroma, and tumour‐infiltrating lymphocytes (TILs). A cutoff of >44.4 TILs per mm2 carcinoma gave a sensitivity of 88% and a specificity of 73% in classifying MMRD carcinomas. Algorithm measures of tumour budding (TB) and poorly differentiated clusters (PDCs) outperformed TB grade derived from routine sign‐out, and compared favourably with manual counts of TB/PDCs with regard to lymphatic, venous and perineural invasion. Comparable associations were seen between algorithm measures of TB/PDCs and manual counts of TB/PDCs for lymph node metastasis (all P < 0.001); however, stronger correlations were seen between the proportion of positive lymph nodes and algorithm measures of TB/PDCs. Stronger associations were also seen between distant metastasis and algorithm measures of TB/PDCs (P = 0.004) than between distant metastasis and TB (P = 0.04) and TB/PDC counts (P = 0.06).ConclusionsOur results highlight the potential of deep learning to identify and quantify a broad spectrum of histological features in colorectal carcinoma.