Objective: To analyze risk and patterns of locoregional failure (LRF) in patients of the RAPIDO trial at 5 years. Background: Multimodality treatment improves local control in rectal cancer. Total neoadjuvant treatment (TNT) aims to improve systemic control while local control is maintained. At 3 years, LRF rate was comparable between TNT and chemoradiotherapy in the RAPIDO trial. Methods: A total of 920 patients were randomized between an experimental (EXP, short-course radiotherapy, chemotherapy, and surgery) and a standard-care group (STD, chemoradiotherapy, surgery, and optional postoperative chemotherapy). LRFs, including early LRF (no resection except for organ preservation/R2 resection) and locoregional recurrence (LRR) after an R0/R1 resection, were analyzed. Results: Totally, 460 EXP and 446 STD patients were eligible. At 5.6 years (median follow-up), LRF was detected in 54/460 (12%) and 36/446 (8%) patients in the EXP and STD groups, respectively (P=0.07), in which EXP patients were more often treated with 3-dimensional-conformed radiotherapy (P=0.029). In the EXP group, LRR was detected more often [44/431 (10%) vs. 26/428 (6%); P=0.027], with more often a breached mesorectum (9/44 (21%) vs. 1/26 (4); P=0.048). The EXP treatment, enlarged lateral lymph nodes, positive circumferential resection margin, tumor deposits, and node positivity at pathology were the significant predictors for developing LRR. Location of the LRRs was similar between groups. Overall survival after LRF was comparable [hazard ratio: 0.76 (95% CI, 0.46–1.26); P=0.29]. Conclusions: The EXP treatment was associated with an increased risk of LRR, whereas the reduction in disease-related treatment failure and distant metastases remained after 5 years. Further refinement of the TNT in rectal cancer is mandated.

Cox Proportional Hazards (CPH) analysis is the standard for survival analysis in oncology. Recently, several machine learning (ML) techniques have been adapted for this task. Although they have shown to yield results at least as good as classical methods, they are often disregarded because of their lack of transparency and little to no explainability, which are key for their adoption in clinical settings. In this paper, we used data from the Netherlands Cancer Registry of 36,658 non-metastatic breast cancer patients to compare the performance of CPH with ML techniques (Random Survival Forests, Survival Support Vector Machines, and Extreme Gradient Boosting [XGB]) in predicting survival using the $$c$$c -index. We demonstrated that in our dataset, ML-based models can perform at least as good as the classical CPH regression ( $$c$$c -index $$\sim \,0.63$$∼0.63 ), and in the case of XGB even better ( $$c$$c -index $$\sim 0.73$$∼0.73 ). Furthermore, we used Shapley Additive Explanation (SHAP) values to explain the models’ predictions. We concluded that the difference in performance can be attributed to XGB’s ability to model nonlinearities and complex interactions. We also investigated the impact of specific features on the models’ predictions as well as their corresponding insights. Lastly, we showed that explainable ML can generate explicit knowledge of how models make their predictions, which is crucial in increasing the trust and adoption of innovative ML techniques in oncology and healthcare overall.

Purpose: Patients with rare cancers (incidence less than 6 cases per 100,000 persons per year) commonly have less treatment opportunities and are understudied at the level of genomic targets. We hypothesized that patients with rare cancer benefit from approved anticancer drugs outside their label similar to common cancers. Experimental Design: In the Drug Rediscovery Protocol (DRUP), patients with therapy-refractory metastatic cancers harboring an actionable molecular profile are matched to FDA/European Medicines Agency–approved targeted therapy or immunotherapy. Patients are enrolled in parallel cohorts based on the histologic tumor type, molecular profile and study drug. Primary endpoint is clinical benefit (complete response, partial response, stable disease ≥ 16 weeks). Results: Of 1,145 submitted cases, 500 patients, including 164 patients with rare cancers, started one of the 25 available drugs and were evaluable for treatment outcome. The overall clinical benefit rate was 33% in both the rare cancer and nonrare cancer subgroup. Inactivating alterations of CDKN2A and activating BRAF aberrations were overrepresented in patients with rare cancer compared with nonrare cancers, resulting in more matches to CDK4/6 inhibitors (14% vs. 4%; P ≤ 0.001) or BRAF inhibitors (9% vs. 1%; P ≤ 0.001). Patients with rare cancer treated with small-molecule inhibitors targeting BRAF experienced higher rates of clinical benefit (75%) than the nonrare cancer subgroup. Conclusions: Comprehensive molecular testing in patients with rare cancers may identify treatment opportunities and clinical benefit similar to patients with common cancers. Our findings highlight the importance of access to broad molecular diagnostics to ensure equal treatment opportunities for all patients with cancer.

In contrast to primary colorectal cancer (CRC) little is known about the genomic landscape of metastasized CRC. Here we present whole genome sequencing data of metastases of 429 CRC patients participating in the pan-cancer CPCT-02 study (NCT01855477). Unsupervised clustering using mutational signature patterns highlights three major patient groups characterized by signatures known from primary CRC, signatures associated with received prior treatments, and metastasis-specific signatures. Compared to primary CRC, we identify additional putative (non-coding) driver genes and increased frequencies in driver gene mutations. In addition, we identify specific genes preferentially affected by microsatellite instability. CRC-specific 1kb-10Mb deletions, enriched for common fragile sites, and LINC00672 mutations are associated with response to treatment in general, whereas FBXW7 mutations predict poor response specifically to EGFR-targeted treatment. In conclusion, the genomic landscape of mCRC shows defined changes compared to primary CRC, is affected by prior treatments and contains features with potential clinical relevance.

Hand–foot syndrome (HFS) is a common side-effect of capecitabine. S-1 is an oral fluoropyrimidine with comparable efficacy to capecitabine in gastrointestinal cancers but associated with a lower incidence of HFS in Asian patients. This study compares the incidence of HFS between S-1 and capecitabine as first-line treatment in Western metastatic colorectal cancer (mCRC) patients. Patients with previously untreated mCRC and planned treatment with fluoropyrimidine monochemotherapy were randomized 1 : 1 to receive either capecitabine (1250 mg/m² orally for patients <70 years; 1000 mg/m² for patients ≥70 years, twice daily on days 1–14) or S-1 (30 mg/m² orally twice daily on days 1–14) in 3-weekly cycles, with bevacizumab optional in both groups. The primary endpoint was the incidence of any grade HFS, as assessed by both physicians and patients (diaries). Secondary endpoints included grade 3 HFS, other toxicities, relative dose intensity, progression-free survival, response rate and overall survival. A total of 161 patients were randomized in 27 centres. The incidence of any grade HFS as assessed by physicians was 73% in the capecitabine group (n = 80) and 45% in the S-1 group (n = 80) [odds ratio (95% confidence interval) 0.31 (0.16–0.60), P = 0.0005]. The incidence of grade 3 HFS was 21% and 4% (P = 0.003), respectively. Patient-assessed any grade HFS was 84% and 58%, respectively (P = 0.004). Grade 3 anorexia was more common in the S-1 group (3% versus 13%, P = 0.03). Median relative dose intensity was 88% in the capecitabine group and 95% in the S-1 group (P = 0.026). There were no statistically significant differences in median progression-free survival, response rate and overall survival rates. Treatment with S-1 in Western mCRC patients is associated with a significantly lower incidence of HFS compared with capecitabine, with comparable efficacy. NCT01918852.