A phase II trial of mTORC1/2 inhibition in STK11 deficient non small cell lung cancer

Abstract

There are no current stratified medicine options for STK11-deficient NSCLC. STK11 loss mediates mTORC activation, GLUT1 up-regulation and increased glycolysis. This metabolic reprogramming might represent a therapeutic vulnerability targetable with mTORC1/2 inhibition. In arm B2 of the National Lung Matrix Trial 54 patients with NSCLC received vistusertib, of which 49 were STK11-deficient (30 with KRAS mutation (B2D), 19 without (B2S)). Objective response (OR) and durable clinical benefit (DCB) rates with 95% credible intervals (CrI) were estimated from posterior probability distributions generated using Bayesian beta-binomial conjugate analysis. In B2D, 2 per-protocol patients obtained OR (estimated true OR rate (95%CrI) 9.8% (2.4-24.3). Estimates of true DCB rate (95%CrI): B2D 24.4% (11.1-42.3), B2S 14.6% (3.6-34.7). Overall, vistusertib cannot be recommended in this context. Longitudinal ctDNA analysis demonstrates enrichment of SMARCA4 mutations post-treatment. In vitro studies show adaptive resistance to mTORC1/2 inhibition via AKT reactivation. (NCT02664935, ISRCTN38344105, EudraCT 2014-000814-73, 10 June 2015).

Fig. 1

Patient flow diagram for Arm B of the National Lung Matrix Trial.

Fig. 2. Swimmer plot for participants with STK11+/- KRAS (Cohort B2) as indicated, ordered by progression-free survival time.

CT assessments are indicated and coded by Complete or Partial Response, Stable Disease, Progressive Disease or Not Evaluable according to RECIST v1.1. Follow-up time after treatment discontinuation, commencing a post-trial treatment, and death are also indicated. The solid black vertical line indicates 3 months, and the solid red line 22 weeks, being the minimum time that a participant can be counted as reaching the Durable Clinical Benefit threshold.

Fig. 3. Summarised clinical endpoints (objective response rate, durable clinical benefit rate, progression-free survival, overall survival).

Forest plots indicating the median and 95% Credible Interval for (A) Objective Response rate, (B) Durable Clinical Benefit rate, (C) median Progression-free Survival time, and (D) median Overall Survival time. The vertical red lines for the co-primary outcomes of OR and DCB indicate the minimum clinically relevant threshold for the true rates that would generate a GO decision. Kaplan-Meier curves for (E) Progression-free Survival time, and (F) Overall Survival time.

Fig. 4. Waterfall plot indicating the Best Percentage Change in sum of target lesion diameters for patients in the B2 cohorts.

Per-protocol participants who discontinued without having returned a measurement are included on the left of the plot with a default measurement of +100%.

Fig. 5. Cell viability of STK11-mutant lung cancer cell lines treated with vistusertib (mTORC1/2 inhibition) or everolimus (mTORC1-only inhibition).

A–D Cell viability of NCI-H460, A-427, NCI-A549 (STK11mut KRASmut) and ChaGoK1, NCI-H1755, CAL-12T (STK11mut KRASwt) after treatment with 0 - 10 µM vistusertib or everolimus for 48 hours, expressed as mean ± SEM from at least 3 independent experiments. Pharmacologically relevant concentrations are indicated in shaded area, calculated from previous publications^,. E Relative luminescence for cells treated with 1µM everolimus vs. 1µM vistusertib (data presented from same experiements as described in A–D). F shows calculated IC50 values for everolimus and vistusertib in STK11 mutant NSCLC cell lines.

Fig. 6. Effect of vistusertib (AZD2014) on cell signalling.

MCF7 (control breast cancer cell line), NCI-A549 and NCI-H460 (STK11^MUTKRAS^MUT) (left) and MCF7, ChaGo-K-1 and CAL-12T (STK11^MUTKRAS^WT) (right) treated with 2 µM vistusertib. Lysates were collected at indicated times. Experiment performed three times on independent biological replicates, representative blots are shown.