Background: The field of liquid biopsy provides a promising alternative to standard tissue biopsies. Previous work has shown that plasma circulating cell-free DNA (ctDNA) can reflect the heterogeneous spectrum of mutations in cancer including metastatic melanoma. Our project aimed to establish and statistically validate plasma-based assays for tumour load and therapy monitoring in melanoma.
Methods: On a large cohort of stage III and stage IV melanoma patients (N = 96) who received signalling targeted or immune checkpoint inhibitors we showed that the most common oncogenic drivers of this disease such as the BRAFV600E, NRASQ61 and the TERTC250T and TERTC228T promoter mutations (termed TERTprom) can be analysed in ctDNA with highly sensitive droplet digital PCR technology (detection of mutant ctDNA down to 0.01% analytical sensitivity).
Results: Our research has demonstrated that ctDNA (irrespective of the genotype) significantly correlates with tumour stage (P < 0.05). Using receiver operating characteristics (ROC) analyses thresholds were established for risk stratification and response prediction. Elevated ctDNA at baseline was a significant predictor of disease progression compared to elevated LDH or S100 in multivariable cox proportional hazards model (Hazard ratio [HR] 7.43, P = 0.05). During therapy, patients with low ctDNA load (below the ROC threshold) had significantly better radiological outcomes and prolonged progression free survival (PFS) compared to patients with high ctDNA load (P < 0.0001). Our findings were confirmed on an independent cohort of metastatic melanoma patients (N = 35) treated with immune checkpoint inhibitors, where also during therapy low ctDNA load correlated with prolonged PFS (P = 0.003). An added benefit of ctDNA was demonstrated in about 80% of the patients, where ctDNA analyses preceded the radiological diagnosis of response or relapse. Progression was detected in plasma ctDNA in average 3.5 months earlier as compared to routine imaging techniques. Finally, we demonstrated that the occurrence of NRASQ61 mutation in BRAFV600-inhibitor treated patients at therapy baseline was associated with treatment failure. The sub-clonal NRASQ61 mutation at therapy baseline was an independent predictor of shorter PFS (HR 2.69, P = 0.02) as compared to BRAFV600E patients without the NRASQ61 mutation at therapy baseline.
Conclusions: In sum, our results support the value of ctDNA as a sensitive biomarker for real-time therapy monitoring and early detection of disease progression.