Explainable machine learning via intra-tumoral radiomics feature mapping for patient stratification in adjuvant chemotherapy for locoregionally advanced nasopharyngeal carcinoma.

PURPOSE: This study aimed to discover intra-tumor heterogeneity signature and validate its predictive value for adjuvant chemotherapy (ACT) following concurrent chemoradiotherapy (CCRT) in locoregionally advanced nasopharyngeal carcinoma (LA-NPC). MATERIALS AND METHODS: 397 LA-NPC patients were retrospectively enrolled. Pre-treatment contrast-enhanced T1-weighted (CET1-w) MR images, clinical variables, and follow-up were retrospectively collected. We identified single predictive radiomic feature from primary gross tumor volume (GTVnp) and defined predicted subvolume by calculating voxel-wised feature mapping and within GTVnp. We independently validate predictive value of identified feature and associated predicted subvolume. RESULTS: Only one radiomic feature, gldm_DependenceVariance in 3 mm-sigma LoG-filtered image, was discovered as a signature. In the high-risk group determined by the signature, patients received CCRT + ACT achieved 3-year disease free survival (DFS) rate of 90% versus 57% (HR, 0.20; 95%CI, 0.05-0.94; P = 0.007) for CCRT alone. The multivariate analysis showed patients receiving CCRT + ACT had a HR of 0.21 (95%CI: 0.06-0.68, P = 0.009) for DFS compared to those receiving CCRT alone. The predictive value can also be generalized to the subvolume with multivariate HR of 0.27 (P = 0.017) for DFS. CONCLUSION: The signature with its heterogeneity mapping could be a reliable and explainable ACT decision-making tool in clinical practice.

Dynamic Changes of Post-Radiotherapy Plasma Epstein-Barr Virus DNA in a Randomized Trial of Adjuvant Chemotherapy Versus Observation in Nasopharyngeal Cancer.

PURPOSE: To study the dynamic changes in plasma Epstein-Barr virus (pEBV) DNA after radiotherapy in nasopharyngeal cancer (NPC). EXPERIMENTAL DESIGN: We conducted a randomized controlled trial of adjuvant chemotherapy versus observation in patients with NPC who had detectable pEBV DNA at 6 weeks post-radiotherapy. Randomized patients had a second pEBV DNA checked at 6 months post-randomization. The primary endpoint was progression-free survival (PFS). RESULTS: We prospectively enrolled 789 patients. Baseline post-radiotherapy pEBV DNA was undetectable in 573 (72.6%) patients, and detectable in 216 (27.4%) patients, of whom 104 (13.2%) patients were eligible for randomization to adjuvant chemotherapy (n = 52) versus observation (n = 52). The first post-radiotherapy pEBV DNA had a sensitivity of 0.48, specificity of 0.81, area under receiver-operator characteristics curve (AUC) of 0.65, false positive (FP) rate of 13.8%, and false negative (FN) rate of 14.4% for disease progression. The second post-radiotherapy pEBV DNA had improved sensitivity of 0.81, specificity of 0.75, AUC of 0.78, FP rate of 14.3%, and FN rate of 8.1%. Patients with complete clearance of post-radiotherapy pEBV DNA (51%) had survival superior to that of patients without post-radiotherapy pEBV DNA clearance (5-year PFS, 85.5% vs. 23.3%; HR, 9.6; P < 0.0001), comparable with patients with initially undetectable post-radiotherapy pEBV DNA (5-year PFS, 77.1%), irrespective of adjuvant chemotherapy or observation. CONCLUSIONS: Patients with NPC with detectable post-radiotherapy pEBV DNA who experienced subsequent pEBV DNA clearance had superior survival comparable with patients with initially undetectable post-radiotherapy pEBV DNA. Post-radiotherapy pEBV DNA clearance may serve as an early surrogate endpoint for long-term survival in NPC.