Assessing tumor heterogeneity using ctDNA to predict and monitor therapeutic response in metastatic breast cancer

Tumor heterogeneity was associated with treatment outcome of metastatic cancers but few studies have examined whether tumor heterogeneity in circulating tumor DNA (ctDNA) can be used to predict treatment outcome. ctDNA analysis was performed in 37 HER2-positive metastatic breast cancer patients treated with pyrotinib. Patients with high tumor heterogeneity had significantly worse PFS outcomes, with a median PFS of 30.0 weeks vs. 60.0 weeks for patients with low tumor heterogeneity (hazard ratio [HR], 2.9; p = 0.02). Patients with trunk resistance mutations receiving pyrotinib monotherapy had worse outcomes (HR, 4.5; p = 0.03), with a median PFS of 7.8 weeks vs. 27.4 weeks for those with branch resistance mutations or without any resistance mutations in baseline ctDNA. Longitudinal monitoring of 21 patients during treatment showed that the molecular tumor burden index ([mTBI] a measure of the percentage of ctDNA in samples) was positively correlated with tumor size as evaluated by computed tomography (p < 0.0001, Pearson r = 0.52) and detected disease progression 8-16 weeks earlier. Our current findings suggested that ctDNA could be used to assess tumor heterogeneity and predict treatment outcomes. Furthermore, the mTBI is better for assessing therapeutic response than single gene mutations and might supplement the current therapeutic response evaluation system.