Abstract 856: Number of ALK-amplified circulating tumor cells predicts progression-free survival in ALK-rearranged non-small cell lung cancer patients treated by crizotinib

The duration of clinical response is unpredictable in ALK-rearranged NSCLC patients treated by crizotinib but all patients invariably develop resistance. Using filter-adapted fluorescence in situ hybridization (FA-FISH) we previously reported ALK-rearrangement detection in circulating tumor cells (CTCs) from ALK-rearranged patients. Here we report the monitoring of ALK-rearranged (≥1×3’/5’, ≥1x(3’ and 5’)) and ALK-amplified (>2×3’/5’) CTCs at baseline and at an early time point under crizotinib therapy in an extended cohort of ALK-rearranged patients. The correlation between CTC subsets harboring distinct FISH patterns and clinical parameters including progression-free survival (PFS) and overall survival (OS) is presented. Forty ALK-rearranged patients were recruited. Blood samples were collected at baseline to crizotinib and at 1-3 months. Abnormal ALK-FISH patterns were examined in CTCs using immunofluorescence staining (DAPI/CD45) combined with FA-FISH after isolation by size of epithelial tumor cells (ISET) filtration. CTCs were defined into five distinct subsets according to the number of FISH ‘break-apart’ signals (3’ and 5’) or isolated red signals (3’) and/or native copies (3’/5’). All ALK-abnormal cells were validated by a cytogenetician. Clinical data was collected retrospectively. Confirming our previous data, ALK rearrangement was detected in CTCs: 32/39 patients (82%) had ≥4 ALK-rearranged CTCs per 1 ml of blood. Age >60 years and smoking status >15 pack years were associated with a higher mean number of ALK-amplified CTCs (p = 0.0423 and p = 0.0407 respectively). No statistically significant correlation was observed between the different CTC subsets with abnormal FISH patterns at baseline and PFS or OS. However we observed a statistically significant correlation (p = 0.0196) between the evolution under crizotinib of the numbers of ALK-amplified CTCs and PFS. An increase in the numbers of ALK-amplified CTCs during treatment is associated with a PFS of 6.5 months, while a decrease in this subset is associated with a PFS of 25 months. There was no corresponding correlation with OS. The evolution under crizotinib therapy of numbers of ALK-amplified CTCs is significantly correlated with PFS. Although not dominant, amplification of the ALK gene has been reported to be one mechanism of acquired resistance to crizotinib therapy in tumor biopsies. CTCs with an amplification of ALK do not express the oncogenic ALK fusion protein and are therefore not targeted by crizotinib. FISH patterns suggested these CTCs may have a high degree of ploidy and chromosomal instability which may contribute to promote the emergence of CTC subclones with a high metastatic potential. Our data suggest that number of ALK-amplified CTCs may be a predictive biomarker allowing prediction of ALK-rearranged NSCLC patients who are at risk of early resistance to crizotinib. Citation Format: Emma Pailler, Marianne Oulhen, Kirsty Ross, Fanny Billiot, Nathalie Auger, Isabelle Borger, Maud NgoCamus, Jordi Remon-Masip, David Planchard, Jean-Charles Soria, Benjamin Besse, Françoise Farace. Number of ALK-amplified circulating tumor cells predicts progression-free survival in ALK-rearranged non-small cell lung cancer patients treated by crizotinib. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 856.