Background: To explore the potential value of consensus molecular subtypes (CMS) in stage II colon cancer treatment selection, we carried out an early cost-effectiveness assessment of a CMS-based strategy for adjuvant chemotherapy. Methods: We used a Markov cohort model to evaluate three selection strategies: (i) the Dutch guideline strategy (MSS+pT4), (ii) the mutation-based strategy (MSS plus a BRAF and/or KRAS mutation or MSS plus pT4), and (iii) the CMS-based strategy (CMS4 or pT4). Outcomes were number of colon cancer deaths per 1,000 patients, total discounted costs per patient (pp), and quality-adjusted life-years (QALY) pp. The analyses were conducted from a Dutch societal perspective. The robustness of model predictions was assessed in sensitivity analyses. To evaluate the value of future research, we performed a value of information (VOI) analysis. Results: The Dutch guideline strategy resulted in 8.10 QALYs pp and total costs of €23,660 pp. The CMS-based and mutation-based strategies were more effective and more costly, with 8.12 and 8.13 QALYs pp and €24,643 and €24,542 pp, respectively. Assuming a threshold of €50,000/QALY, the mutation-based strategy was considered as the optimal strategy in an incremental analysis. However, the VOI analysis showed substantial decision uncertainty driven by the molecular markers (expected value of partial perfect information: €18M). Conclusions: On the basis of current evidence, our analyses suggest that the mutation-based selection strategy would be the best use of resources. However, the extensive decision uncertainty for the molecular markers does not allow selection of an optimal strategy at present. Impact: Future research is needed to eliminate decision uncertainty driven by molecular markers.

中文翻译：

---

II期结肠癌辅助治疗基于风险的选择策略的早期成本效益分析：预后分子标志物的潜在价值

背景：为了探索共识分子亚型 (CMS) 在 II 期结肠癌治疗选择中的潜在价值，我们对基于 CMS 的辅助化疗策略进行了早期成本效益评估。方法：我们使用马尔科夫队列模型评估三种选择策略：（i）荷兰指南策略（MSS+pT4），（ii）基于突变的策略（MSS 加 BRAF 和/或 KRAS 突变或 MSS 加 pT4） ，以及 (iii) 基于 CMS 的策略（CMS4 或 pT4）。结果是每 1,000 名患者的结肠癌死亡人数、每位患者的总贴现成本 (pp) 和质量调整生命年 (QALY) pp。这些分析是从荷兰社会的角度进行的。在敏感性分析中评估了模型预测的稳健性。为了评估未来研究的价值，我们进行了信息价值（VOI）分析。结果：荷兰指南策略产生 8.10 QALYs pp 和 23,660 欧元 pp 的总成本。基于 CMS 和基于突变的策略更有效且成本更高，分别为 8.12 和 8.13 QALYs pp 以及 24,643 欧元和 24,542 欧元 pp . 假设阈值为 50,000 欧元/QALY，基于突变的策略被认为是增量分析中的最佳策略。然而，VOI 分析显示出由分子标记驱动的重大决策不确定性（部分完美信息的预期价值：1800 万欧元）。结论：根据目前的证据，我们的分析表明基于突变的选择策略将是资源的最佳利用。然而，分子标记的广泛决策不确定性目前不允许选择最佳策略。影响：未来的研究需要消除由分子标记驱动的决策不确定性。