Kinase inhibitors can be highly effective against cancers driven by specific oncogenic kinase proteins, but resistance usually develops after prolonged treatment. Target-dependent mechanisms, typically acquired second-site mutations that prevent drug binding, or target-independent mechanisms, such as downstream or parallel pathway activation, may mediate resistance. Prolonged exposure of kinase-dependent cell lines to inhibitors in vitro to select for resistant populations has identified clinically relevant resistance mechanisms against a number of inhibitors. ALK-positive anaplastic large-cell lymphoma (ALCL) is an uncommon form of peripheral T-cell lymphoma (PTCL) driven by the NPM-ALK fusion kinase, resulting from a t(2;5)(p23;q35) chromosomal rearrangement. ALK inhibitors developed for lung cancer, where a different ALK fusion (EML4-ALK) is found in 6% of cases, have shown promise clinically as a new treatment modality for ALK+ ALCL patients. In this study, we selected three NPM-ALK-dependent ALCL cell lines (Karpas-299, SUP-M2, and SU-DHL-1) for resistance to the ALK inhibitor crizotinib. All three lines initially were highly sensitive the drug (IC50< 100 nM) and required months of exposure at gradually increasing concentrations before acquiring resistance. Previous work has identified specific ALK kinase second-site mutations that promote crizotinib resistance, which we also observed in some resistant lines generated. A unique situation arose, however, in a sub-clone of Karpas-299, that was able to grow in crizotinib at concentrations > 1 µM (Karpas-299CR1000). This line had greatly increased viability in crizotinib at concentrations up to 1 µM, compared to untreated cells, suggesting the drug actually maintained growth. Plated in 250 nM crizotinib, Karpas-299CR1000 grows rapidly, similar to the parent line growing in drug-free media. Plated in drug-free media, however, the line’s viability drops rapidly to zero and dies, similar to the parent line plated in 250 nM crizotinib. Therefore the drug that once shut off the signaling necessary for the cells’ survival, after prolonged exposure (>9 months), became necessary to maintain it in the sub-clonal population. Melanoma cells driven by BRAF-V600E selected for vemurafenib resistance had a similar phenotype in a recent report (Das Thakur, et al., 2013). These cells amplified mutant BRAF to overcome the inhibitor but came to require the drug to counteract BRAF over-activity with amplification beyond a certain point. Preliminarily, we find a similar effect of NPM-ALK amplification in Karpas-299CR1000 cells but are actively exploring mechanism. In sum, our study shows that long-term exposure of ALK+ ALCL cells to an ALK inhibitor may induce a resistant phenotype that comes to depend on drug presence to prevent over-activity of the mutant kinase.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.