Liu G, Huang S H, Ailles L, Rey-McIntyre K, Melton C A, Shen S Y, Burgener J M, Brown B, Zhang J, Min J, Wang Y, Hall O, Jones J T, Budhraja K, Provance J B, Sosa E V, Licon A, Williams A, Bratman S V, Allen B A, Zhang J, Hartman A-R, De Carvalho D D
Princess Margaret Cancer Centre, University Health Network, Toronto; Department of Medical Biophysics, University of Toronto, Toronto, Canada.
Princess Margaret Cancer Centre, University Health Network, Toronto.
Ann Oncol. 2025 Jan;36(1):108-117. doi: 10.1016/j.annonc.2024.08.2348. Epub 2024 Sep 14.
Outcomes for patients with locally advanced head and neck cancer (HNC) treated with curative intent remain disappointing, with 5-year survival rates at 50%. Most recurrences occur within the first 2 years after treatment, providing a window of opportunity to identify patients with molecular residual disease (MRD). A tissue-agnostic test for MRD detection in patients with human papillomavirus (HPV) positive and negative HNC, where tissue is often scarce, is needed.
Patients with stage I-IVB HNC, including patients positive and negative for HPV, were enrolled and peripheral blood plasma was collected longitudinally at diagnosis and ∼3, 12, and 24 months after curative intent treatment. The full cohort includes 325 patients with 1155 samples. Samples were split into distinct sets to train and validate a classifier capable of identifying MRD using a tissue-agnostic genome-wide methylome enrichment platform. The primary endpoint was recurrence-free survival (RFS).
With a median follow-up of 60 months, patients in the blinded validation set with MRD positivity experienced significantly worse RFS with a hazard ratio (HR) of 35.7 [95% confidence interval (CI) 10.8-117.8; P < 0.0001]. For patients with HPV negativity, HR was 42.3 (95% CI 9.8-182.3; P < 0.0001); for patients with HPV-positive oropharyngeal cancer, HR was 24.1 (95% CI 3.0-196.8; P < 0.0001). Moreover, the lead time between MRD positivity and clinical recurrence was up to 14.9 months, with a mean lead time of 4.1 months. Surveillance sensitivity was 91% (95% CI 77% to 97%) and specificity was 88% (95% CI 80% to 93%).
Here we validate the clinical performance characteristics of a tissue-agnostic genome-wide methylome enrichment assay for MRD detection in patients with HNC. The MRD detection test showed high sensitivity for identifying recurrence at high specificity across different anatomical sites, HPV status, and treatment regimens, highlighting the broad applicability for MRD detection in patients with HNC.
接受根治性治疗的局部晚期头颈癌（HNC）患者的预后仍然令人失望，5年生存率为50%。大多数复发发生在治疗后的前2年内，这为识别分子残留疾病（MRD）患者提供了一个机会窗口。对于人乳头瘤病毒（HPV）阳性和阴性的HNC患者，由于组织往往稀缺，需要一种不依赖组织的MRD检测方法。
纳入I-IVB期HNC患者，包括HPV阳性和阴性患者，在诊断时以及根治性治疗后约3、12和24个月纵向采集外周血血浆。整个队列包括325例患者和1155份样本。样本被分成不同的组，以训练和验证一种能够使用不依赖组织的全基因组甲基化富集平台识别MRD的分类器。主要终点是无复发生存期（RFS）。
中位随访60个月，盲法验证组中MRD阳性的患者RFS明显更差，风险比（HR）为35.7[95%置信区间（CI）10.8-117.8；P<0.0001]。对于HPV阴性患者，HR为42.3（95%CI 9.8-182.3；P<0.0001）；对于HPV阳性口咽癌患者，HR为24.1（95%CI 3.0-196.8；P<0.0001）。此外，MRD阳性与临床复发之间的提前期长达14.9个月，平均提前期为4.1个月。监测敏感性为91%（95%CI 77%至97%），特异性为88%（95%CI 80%至93%））。
在此，我们验证了一种不依赖组织的全基因组甲基化富集检测方法在HNC患者中检测MRD的临床性能特征。MRD检测试验在不同解剖部位、HPV状态和治疗方案中均显示出高特异性识别复发的高敏感性，突出了其在HNC患者中检测MRD的广泛适用性。
