Interview: Rafał Dziadziuszko, MD, PhD, Department of Oncology and Radiotherapy, Medical University of Gdańsk, Gdańsk, Poland What can we expect from circulating free DNA (cfDNA) as a biomarker in the setting of lung cancer diagnosis and treatment today? Blood-based diagnostics can be used in the field of diagnosis of lung cancer, but also for the evaluation of predictive molecular alterations. Today, lung cancer is divided into many small subsets of patients according to individual aberrations in their DNA. With modern technologies, we can diagnose these alterations not only in tissue, but also in the patient’s blood. It is extremely interesting to see that the diagnostic accuracy of blood-based tests is improving and can be as high as 80 % or even 90 %. These tests can be used for the initial diagnosis and the selection of the targeted agent, including immunotherapies, and potentially also for treatment monitoring. While tests of circulating free DNA (cfDNA) allow for a general assessment of the presence of DNA in the plasma, circulating tumor DNA (ctDNA) focuses on specific alterations with the possibility of establishing allele frequency and quantifying them. At the ASCO Congress, many presentations focused on both cfDNA and ctDNA in patients treated with targeted therapies or immunotherapy, also for diagnostic purposes including screening for the presence of lung cancer, which is becoming a reality. Where do you see the clinical significance of cfDNA in ALK-positive tumors?ALK-positive NSCLC represents 5 % of lung adenocarcinomas. This is a clinically important subset of patients, as they can be treated with ALK inhibitors and enjoy unprecedented survival of several years or more. ALK rearrangements can be diagnosed not only in the tissue, but also in the plasma, again with diagnostic accuracy of approximately 80 %. Moreover, it is possible to detect variants of ALK translocations and quantify the amount of circulating ALK in the blood. This has been shown to correlate with response. Decreases in the ALK allele frequency in the course of treatment usually tell us that the treatment will be effective. As for other markers, diagnosis and monitoring for treatment efficacy in clinical trials investigating ALK inhibitors can be based on blood testing. What are the results of the analysis you presented at this year’s ASCO Congress? This year I had the pleasure of representing the ALEX investigators who contributed to the enrollment of patients with ALK-positive NSCLC . Patients were randomized to first-line therapy with either crizotinib, which used to be the standard of care, or the novel ALK inhibitor alectinib. PFS, OS and other endpoints were assessed. The analysis we presented investigated circulating free tumor DNA as a proxy of tumor burden and correlated that with patient prognosis. We showed that cfDNA correlates with tumor burden; patients with a high number of metastases or huge tumors had higher cfDNA levels and worse prognosis with both crizotinib and alectinib, although alectinib-treated patients did better independent of their cfDNA levels. Therefore, cfDNA probably explains why PFS and OS differ across patients. In addition, it should be noted that cfDNA testing is quite a simple measurement. Out of 300 ALEX patients, the data for over 270 individuals were available for the statistical analysis. At present, these results only have prognostic implications. The underlying tumor burden that is quantified using a simple test can provide information on which patient will do well and which patient has a higher likelihood of failing on ALK-targeted treatment. In the future, however, we would also like to evaluate the temporal changes in cfDNA and investigate associations with radiologic relapse. Also, cfDNA tests might be used as predictive assays to identify patients who do not initially respond to ALK inhibitor therapy. This remains to be explored.
Dziadziuszko R et al., Circulating free DNA as a prognostic biomarker in patients with advanced ALK+ NSCLC treated with alectinib from the global phase III ALEX trial. J Clin Oncol 37, 2019 (suppl; abstr 9053)