Henry Walch

BACKGROUND: The causative factors for the recent increase in early-onset colorectal cancer (EO-CRC) incidence are unknown. We sought to determine if early-onset disease is clinically or genomically distinct from average-onset colorectal cancer (AO-CRC). METHODS: Clinical, histopathologic, and genomic characteristics of EO-CRC patients (2014-2019), divided into age 35 years and younger and 36-49 years at diagnosis, were compared with AO-CRC (50 years and older). Patients with mismatch repair deficient tumors, CRC-related hereditary syndromes, and inflammatory bowel disease were excluded from all but the germline analysis. All statistical tests were 2-sided. RESULTS: In total, 759 patients with EO-CRC (35 years, n = 151; 36-49 years, n = 608) and AO-CRC (n = 687) were included. Left-sided tumors (35 years and younger = 80.8%; 36-49 years = 83.7%; AO = 63.9%; P < .001 for both comparisons), rectal bleeding (35 years and younger = 41.1%; 36-49 years = 41.0%; AO = 25.9%; P = .001 and P < .001, respectively), and abdominal pain (35 years and younger = 37.1%; 36-49 years = 34.0%; AO = 26.8%; P = .01 and P = .005, respectively) were more common in EO-CRC. Among microsatellite stable tumors, we found no differences in histopathologic tumor characteristics. Initially, differences in TP53 and Receptor Tyrosine Kinase signaling pathway (RTK-RAS)alterations were noted by age. However, on multivariate analysis including somatic gene analysis and tumor sidedness, no statistically significant differences at the gene or pathway level were demonstrated. Among advanced microsatellite stable CRCs, chemotherapy response and survival were equivalent by age cohorts. Pathogenic germline variants were identified in 23.3% of patients 35 years and younger vs 14.1% of AO-CRC (P = .01). CONCLUSIONS: EO-CRCs are more commonly left-sided and present with rectal bleeding and abdominal pain but are otherwise clinically and genomically indistinguishable from AO-CRCs. Aggressive treatment regimens based solely on the age at CRC diagnosis are not warranted.

BACKGROUND AND AIM: Genetic alterations in intrahepatic cholangiocarcinoma (iCCA) are increasingly well characterized, but their impact on outcome and prognosis remains unknown. APPROACH AND RESULTS: This bi-institutional study of patients with confirmed iCCA (n = 412) used targeted next-generation sequencing of primary tumors to define associations among genetic alterations, clinicopathological variables, and outcome. The most common oncogenic alterations were isocitrate dehydrogenase 1 (IDH1; 20%), AT-rich interactive domain-containing protein 1A (20%), tumor protein P53 (TP53; 17%), cyclin-dependent kinase inhibitor 2A (CDKN2A; 15%), breast cancer 1-associated protein 1 (15%), FGFR2 (15%), polybromo 1 (12%), and KRAS (10%). IDH1/2 mutations (mut) were mutually exclusive with FGFR2 fusions, but neither was associated with outcome. For all patients, TP53 (P < 0.0001), KRAS (P = 0.0001), and CDKN2A (P < 0.0001) alterations predicted worse overall survival (OS). These high-risk alterations were enriched in advanced disease but adversely impacted survival across all stages, even when controlling for known correlates of outcome (multifocal disease, lymph node involvement, bile duct type, periductal infiltration). In resected patients (n = 209), TP53mut (HR, 1.82; 95% CI, 1.08-3.06; P = 0.03) and CDKN2A deletions (del; HR, 3.40; 95% CI, 1.95-5.94; P < 0.001) independently predicted shorter OS, as did high-risk clinical variables (multifocal liver disease [P < 0.001]; regional lymph node metastases [P < 0.001]), whereas KRASmut (HR, 1.69; 95% CI, 0.97-2.93; P = 0.06) trended toward statistical significance. The presence of both or neither high-risk clinical or genetic factors represented outcome extremes (median OS, 18.3 vs. 74.2 months; P < 0.001), with high-risk genetic alterations alone (median OS, 38.6 months; 95% CI, 28.8-73.5) or high-risk clinical variables alone (median OS, 37.0 months; 95% CI, 27.6-not available) associated with intermediate outcome. TP53mut, KRASmut, and CDKN2Adel similarly predicted worse outcome in patients with unresectable iCCA. CDKN2Adel tumors with high-risk clinical features were notable for limited survival and no benefit of resection over chemotherapy. CONCLUSIONS: TP53, KRAS, and CDKN2A alterations were independent prognostic factors in iCCA when controlling for clinical and pathologic variables, disease stage, and treatment. Because genetic profiling can be integrated into pretreatment therapeutic decision-making, combining clinical variables with targeted tumor sequencing may identify patient subgroups with poor outcome irrespective of treatment strategy.

The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. Patient data are often stored in unstructured text and siloed datasets. Here we combine natural language processing annotations1,2 with structured medication, patient-reported demographic, tumour registry and tumour genomic data from 24,950 patients at Memorial Sloan Kettering Cancer Center to generate a clinicogenomic, harmonized oncologic real-world dataset (MSK-CHORD). MSK-CHORD includes data for non-small-cell lung (n = 7,809), breast (n = 5,368), colorectal (n = 5,543), prostate (n = 3,211) and pancreatic (n = 3,109) cancers and enables discovery of clinicogenomic relationships not apparent in smaller datasets. Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. By annotating 705,241 radiology reports, MSK-CHORD also uncovers predictors of metastasis to specific organ sites, including a relationship between SETD2 mutation and lower metastatic potential in immunotherapy-treated lung adenocarcinoma corroborated in independent datasets. We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. The resulting data are provided as a public resource for real-world oncologic research. A study generates a clinicogenomics dataset resource, MSK-CHORD, that combines natural language processing-derived clinical annotations with patient medical data from various sources to improve models of cancer outcome.