As reported in The Lancet Oncology, Kleppe et al have developed a risk stratification system using the DoMore-v1-CRC marker and pathologic staging markers that may permit many patients with stage II or III colorectal cancer to avoid adjuvant chemotherapy.
As stated by the investigators, “The DoMore-v1-CRC marker was recently developed using deep learning and conventional hematoxylin and eosin-stained tissue sections and was observed to outperform established molecular and morphological markers of patient outcome after primary colorectal cancer resection. The aim of the present study was to develop a clinical decision support system based on DoMore-v1-CRC and [pathologic] staging markers to facilitate individualized selection of adjuvant treatment.”
Study Details
The investigators estimated cancer-specific survival in subgroups defined by pathologic tumor stage (pT < 4 or pT4), pathologic nodal stage (pN0, pN1, or pN2), number of lymph nodes sampled (≤ 12 or > 12) if not pN2, and DoMore-v1-CRC classification (good, uncertain, or poor prognosis) in a development cohort of 997 patients with stage II or III colorectal cancer considered to have no residual tumor (R0) from two community-based cohorts in Norway and the United Kingdom. The data were used to define three risk groups. The risk stratification scheme was tested in an external validation cohort of 1,075 patients with stage II or III R0 colorectal cancer from the QUASAR 2 trial who had received single-agent capecitabine.
Findings were compared with a guideline-based stratification in the validation cohort, in which: patients with pN0, pT3, and > 12 lymph nodes sampled were classified as low-risk; other pN0 patients (ie, those with pT4 or ≤ 12 lymph nodes sampled) were classified as intermediate-risk together with patients with pN1 and pT< 4; and those with stage III disease with pN2 or pT4 were classified as high-risk.
Key Findings
Among 1,075 patients in the validation cohort, 445 (41%) were low-risk, 339 (32%) were intermediate-risk, and 291 (27%) were high-risk using the new risk stratification system. Hazard ratios for cancer-specific survival were 10.71 (95% confidence interval [CI] = 6.39–17.93, P < .0001) for high-risk vs low-risk patients and 3.06 (95% CI = 1.73–5.42, P = .0001) for intermediate-risk vs low-risk patients. Estimated 3-year cancer-specific survival was 97.2% (95% CI = 95.1%–98.4), 94.8% (95% CI = 91.7%–96.7%), and 77.6% (95% CI = 72.1%–82.1%) for the low-, intermediate-, and high-risk groups, respectively.
On the guideline-based risk stratification system, 13%, 57%, and 30% of validation cohort patients were low-, intermediate-, and high-risk, respectively. Estimated 3-year cancer-specific survival was 96.4% (95% CI = 91.5%–98.5%), 95.9% (95% CI = 93.9%–97.2%), and 79.6% (95% CI = 74.6%–83.8%), respectively. Hazard ratios for cancer-specific survival were 6.12 (95% CI = 3.09–12.10) for high- vs low-risk and 1.13 (95% CI = 0.55–2.32) for intermediate- vs low-risk. The DoMore-v1-CRC marker provided prognostic information beyond the guideline-based risk stratification system (P < .000).
In total, 325 (30%) of the 1,075 patients in the validation cohort, comprising 284 with intermediate risk and 41 with high risk according to the guideline-based risk stratification system, were classified as low-risk by the by the new risk stratification system.
The investigators concluded, “Integrating DoMore-v1-CRC and [pathologic] staging markers provided a clinical decision support system that risk stratifies more accurately than its constituent elements, and identifies substantially more patients with stage II and III colorectal cancer with similarly good prognosis as the low-risk group in current guidelines. Avoiding adjuvant chemotherapy in these patients might be safe, and could reduce morbidity, mortality, and treatment costs.”
Håvard E. Danielsen, PhD, of the Institute for Cancer Genetics and Informatics, Oslo University Hospital, is the corresponding author for The Lancet Oncology article.
Disclosure: The study was supported by The Research Council of Norway. For full disclosures of the study authors, visit thelancet.com.
