Larotrectinib in NTRK-Rearranged Solid Tumors

Members of the tropomyosin receptor tyrosine kinase (TRK) family include TRKA, TRKB, and TRKC (encoded by the genes NTRK1, NTRK2, and NTRK3, respectively). In response to activation by neurotrophin ligands, TRK family members regulate critical and diverse neurologic processes including neuronal plasticity and regeneration¹. Of note, recurrent chromosomal rearrangements involving NTRK1, NTRK2, and NTRK3 have been identified in rare pediatric and adult tumors (including infantile fibrosarcoma and certain salivary gland tumors) and in rare subsets of more common tumors such as non-small cell lung carcinoma (NSCLC), colorectal cancer, and breast cancer². These somatic rearrangements generate chimeric proteins fusing the carboxy-terminus of TRK (including the kinase domain) with the amino-terminus of a partner protein that generally contains oligomerization domains (Figure 1A). These fusion proteins demonstrate ligand-independent constitutive kinase activity and function as oncogenic drivers via activation of pathways promoting cell survival and proliferation, irrespective of tissue of origin.

Figure 1.

(A) Schematic representation of a constitutively-activated TRK fusion protein leading to upregulation of the MAP kinase (MAPK), PI3 kinase (PI3K), and phospholipase C gamma (PLCγ) pathways. (B) Chemical structure of larotrectinib. (C) Computed tomography (CT) imaging of the chest reveals a large lung malignancy (arrow) in a 45-year-old female with NSCLC harboring a NTRK1-SQSTM1 rearrangement. (D) Clinical response (arrow) observed after two treatment cycles with larotrectinib. (E) Chemical structure of LOXO-195. Source of chemical structures in (B) and (E): PubChem (https://pubchem.ncbi.nlm.nih.gov). Images in (C) and (D) are adapted by permission from Springer Nature: Springer Nature, Targeted Oncology, TRK Inhibition: A New Tumor-Agnostic Treatment Strategy, Kummar and Lassen, 2018 https://link.springer.com/article/10.1007%2Fs11523-018-0590-1).

Other transforming gene fusions involving receptor tyrosine kinases (such as those generated by ALK and ROS1 rearrangements in NSCLC) have previously been characterized as tractable therapeutic targets amenable to small molecule inhibition. Larotrectinib (LOXO-101, ARRY-470) is an orally-available, potent, and selective ATP-competitive pan-TRK inhibitor with low nanomolar 50% inhibitory concentrations (Figure 1B). Safety and efficacy of larotrectinib in the first 55 adult and pediatric patients with diverse TRK fusion-positive malignancies treated in any of 3 separate phase 1 / 2 studies were recently reported³. While most early-phase clinical trials of investigational agents focus on a single tumor type, remarkably 17 different cancer diagnoses were represented among these 55 patients. NTRK rearrangements were identified prospectively by either next-generation sequencing or by fluorescence in situ hybridization (FISH). Responses were observed in 75% of patients regardless of patient age, tumor type, TRK family member, or identity of the TRK fusion partner (Figure 1C, D). Furthermore, responses were durable with 71% of responses still ongoing at 12 months. Larotrectinib was generally well-tolerated with no study participants requiring discontinuation of the drug due to adverse treatment-related events.

Based on these findings, the Food and Drug Administration granted accelerated approval to larotrectinib in November 2018 for the treatment of adult and pediatric patients with advanced solid tumors harboring an NTRK gene fusion. Of note, larotrectinib represents the first anti-cancer oral targeted therapy to receive FDA approval based solely on the presence of a specific biomarker and without regard to tissue of origin. The observation of responses to larotrectinib across tumor types was not a foregone conclusion, as other targeted anti-cancer therapies are not uniformly active in multiple tumor types. For instance, BRAF inhibitors are approved for use in advanced BRAF-mutant melanoma but are ineffective in BRAF-mutant colorectal cancer. Other TRK inhibitors such as entrectinib (a multikinase inhibitor that also has activity against ALK- and ROS1-driven NSCLC) are in clinical development.

Acquired drug resistance is a nearly universal observation in oncogene-driven solid tumors treated with targeted molecular therapies. A common mechanism of resistance is the development of secondary point mutations within the drug target that impair binding of the inhibitor or increase the ATP affinity of the kinase (leading to increased kinase activity)⁴. In a perhaps unprecedented drug development approach leveraging knowledge of acquired resistance mutations in analogous targetable receptor tyrosine kinases, a strategy was developed to address anticipated TRK resistance mutations. Using structural insights derived from x-ray crystallography of TRK proteins and in silico modeling of additional structurally distinct TRK inhibitors, a second TRK inhibitor called LOXO-195 was identified as a candidate compound with potential activity against secondary mutations conferring resistance to larotrectinib (Figure 1E)⁵. Molecular modeling predicted that bulky amino acid substitutions in the solvent front or substitutions in the xDFG activation domain that impair binding of larotrectinib would not disrupt binding of LOXO-195. Like larotrectinib, LOXO-195 is an orally-available, selective, and potent pan-TRK inhibitor. The development of next-generation kinase inhibitors with activity against secondary mutations conferring resistance to earlier-generation inhibitors is an established therapeutic strategy in other oncogene-driven tumors (such as EGFR mutant and ALK-rearranged NSCLC). However, this is a process that has generally required years to implement due to the need for molecular characterization of patient-derived tumors at resistance and then leveraging these insights for the development of subsequent pharmacologic strategies. Here, based on prior experience with acquired resistance to other kinase-directed therapies, LOXO-195 was developed nearly in parallel with larotrectinib in anticipation of acquired resistance to larotrectinib in treated patients.

And indeed, acquired resistance to larotrectinib was observed in a subset of study participants. Sampling of tumor and/or plasma from 9 trial participants who developed resistance to larotrectinib (after a response or stable disease lasting at least 6 months) demonstrated the acquisition of secondary NTRK mutations in all 9 cases³. These mutations occurred in the gatekeeper position (TRKA F589L), the solvent front position (TRKA G595R or TRKC G623R), or within the xDFG activation domain (TRKA G667S or TRKC G696A). Two patients who developed resistance to larotrectinib associated with a solvent-front secondary mutation (one with TRKA G595R and one with TRKC G623R) were treated with LOXO-195 under single-patient clinical protocols. Both patients experienced tumor regression on LOXO-195, with one patient’s response lasting 3 months and the other ongoing after 6 months of therapy⁵. A Phase 1 / 2 trial evaluating the safety and preliminary efficacy of LOXO-195 in adults and children with TRK-driven solid tumors refractory to prior TRK inhibitor is currently underway.

Although rare, oncogenic TRK fusions in diverse solid tumors delineate a therapeutically tractable patient population, underscoring the need for careful prospective tumor molecular profiling to identify patients most likely to benefit from TRK-directed therapy. The recent tissue-agnostic FDA approval for larotrectinib highlights the potential of precision oncology, where decisions about a therapeutic approach may be based on relevant biomarkers rather than tissue of origin. In addition, the development of larotrectinib and LOXO-195 highlights an innovative paradigm to expedite drug development by anticipating therapeutic resistance and proactively leveraging rational approaches to overcome resistance.