Targeting G1/S phase cell-cycle genomic alterations and accompanying co-alterations with individualized CDK4/6 inhibitor-based regimens

Abstract

BACKGROUNDAlthough CDK4/6 inhibitors are an established treatment for hormone receptor-positive, HER2-negative metastatic breast cancers, their benefit in other malignancies remains limited.METHODSWe investigated factors associated with clinical outcomes from CDK4/6 inhibitor-based therapy among patients with G1/S phase cell-cycle alterations (CDK4/6 amplifications, CCND1/2/3 amplifications, or CDKN2A/B alterations).RESULTSOverall, 2457 patients with diverse solid tumors that underwent clinical-grade, next-generation sequencing (182-465 genes) and therapy outcome of (non-breast cancer) patients treated with matched CDK4/6 inhibitors were analyzed. G1/S phase cell-cycle alterations occurred in 20.6% (507 of 2457) of patients; 99% of those patients (n = 501) harbored ≥1 characterized co-alteration (median, 4; range, 0-24). In 40 patients with G1/S phase cell-cycle alterations given CDK4/6 inhibitors as part of their regimen, significantly longer median progression-free survival (PFS) was observed when CDK4/6 inhibitor-based therapies matched a larger proportion of tumor alterations, often because CDK4/6 inhibitors were administered together with other drugs that were matched to genomic co-alterations, hence achieving a high matching score (high vs. low [≥50% vs. <50%] matching score, PFS, 6.2 vs. 2.0 months, P < 0.001 [n = 40] [multivariate]) and higher rate of stable disease ≥6 months or an objective response (57% vs. 21%, P = 0.048).CONCLUSIONIn summary, in cell-cycle-altered cancers, matched CDK4/6 inhibitors, as part of an individualized regimen targeting a majority of genomic alterations, was independently associated with longer PFS.TRIAL REGISTRATIONClinicalTrials.gov NCT02478931.FUNDINGJoan and Irwin Jacobs Fund, National Cancer Institute (P30 CA023100, R01 CA226803), and the FDA (R01 FD006334).

Keywords: Cell cycle; Molecular biology; Oncology.

Conflict of interest statement

Conflict of interest: SK serves as a consultant for Foundation Medicine: and receives speakers’ fees from Roche. JKS receives research funding from Novartis Pharmaceuticals, Amgen Pharmaceuticals, and Foundation Medicine; consultant fees from Grand Rounds, Loxo, and Deciphera; and speakers’ fees from Roche. RK receives research funding from Incyte, Genentech, Merck Serono, Pfizer, Sequenom, Foundation Medicine, Guardant Health, Grifols, and Konica Minolta as well as consultant fees from Loxo, X-Biotech, Actuate Therapeutics, Genentech, and NeoMed. She receives speakers’ fees from Roche and has an equity interest in IDbyDNA and CureMatch Inc.

Figures

Figure 1. Consort diagram of patients with alterations in the G1/S phase cell-cycle signaling pathway (n = 507).

Figure 2. Summary of co-alterations observed in tumors harboring CDK4/6 amplifications, CCND1/2/3 amplifications, or CDKN2A/B alterations (n = 507).

Among 507 patients with diverse tumors harboring CDK4/6 amplifications, CCND1/2/3 amplifications, or CDKN2A/B alterations, most patients (99%, n = 501) had ≥1 characterized co-alteration (median, 4; range, 0–24) in tissue NGS. The most common co-alterations were in TP53 (approximately 48% of patients, n = 241), EGFR (17% of patients, n = 87), TERT (16% of patients, n = 82), and KRAS genes (16% of patients, n = 81). Genomic alterations with frequency of ≥1.0% were included.

Figure 3. Progression-free survival among patients with alterations in CCND1/2/3, CDK4/6, and/or CDKN2A/B G1/S phase cell-cycle genes, who received CDK4/6 inhibitor–based therapy (n = 40).

(A) Progression-free survival (PFS) comparison between patients who received CDK4/6 inhibitor–based therapy as part of the combination therapies (n = 31) and patients who received CDK4/6 inhibitor as a single agent (n = 9). Among patients with diverse cancers (n = 40) who received CDK4/6 inhibitor–based therapy, there was no significant difference in PFS between patients who received combination therapy and those who received single agents (combination vs. single agent, 4.6 vs. 2.8 months, P = 0.26). (B) PFS among patients who received CDK4/6 inhibitor–based therapy with a matching score of ≥50% (n = 25) versus those with a matching score of <50% (n = 15). Among patients with diverse cancers (n = 40) who received CDK4/6 inhibitor–based therapy, patients who were treated with a combination of agents with higher matching scores had significantly longer PFS (median PFS for matching score ≥50% vs. <50%, 6.2 vs. 2.0 months, P = 0.001). (C) Response to CDK4/6 inhibitor–based therapies among patients with CCND1/2/3, CDK4/6, and/or CDKN2A/B G1/S phase cell-cycle gene alterations. Comparison between patients who received CDK4/6 inhibitor–based therapy with matching score of ≥50% (n = 23) and patients with matching score of <50% (n = 14). There was a significant difference in achieving stable disease ≥6 months/partial response among patients who received therapy with matching score of ≥50% as compared with those with a matching score of <50 (57% vs. 21%, P = 0.048). (Among 40 patients treated with matched CDK4/6 inhibitor–based therapies, 37 patients were assessable for response.) (D) Overall survival (OS) comparison (n = 40) between patients who received CDK4/6 inhibitor–based therapy with matching score of ≥50% (n = 25) and patients with matching score of <50% (n = 15). Among patients who received CDK4/6 inhibitor–based therapy (n = 40), there was no significant difference in OS between those with a matching score of ≥50% vs. <50% (median OS between matching score ≥ 50% vs. < 50%, 8.3 vs. 5.3 months; P = 0.15). Reverse Kaplan-Meier calculation for A, B, and D revealed no difference between groups, indicating that the median follow-up between groups was similar. MS, matching score; PD, progressive disease; PR, partial response; SD, stable disease.

Figure 4. Examples of responders treated with CDK4/6 inhibitory therapy.

(A) Case 1 (patient ID 269): Forty-three-year-old woman with metastatic high-grade ovarian carcinoma with neuroendocrine features that harbored CDKN2A/B alteration without any genomic co-alterations on the NGS panel of 315 genes and 2 lines of prior therapy demonstrated partial response with single-agent palbociclib lasting 8 months. NGS of tumor showed a single alteration in CDKN2A/B, for which the patient was started on palbociclib. Restaging scan with CT overall showed 30% regression, indicating partial response at the 4-month time point (response by RECIST 1.1). Along with the radiographic response, reduction of tumor marker, CA 125 was seen (CA 125: 328 U/ml down to 50 U/ml [reference range 0–34 U/ml]). (B) Case 2 (patient ID 501): Sixty-eight-year-old man with metastatic gastrointestinal stromal tumor (GIST) with alterations in BRAF V600E, CDKN2A p16INK4a splice site 150+1G > A and LRP1B deletion exon 23 presented after progressing treatment with dabrafenib (BRAF inhibitor) and trametinib (MEK inhibitor) based on underlying BRAF V600E mutation (30). Addition of palbociclib led to partial response lasting 11.3 months. Although progression was seen with a new pulmonary nodule and worsening rectal lesion (left to middle, circle), one of the right lower lung masses appeared to be stable (left to middle, arrow), and thus the decision was made to continue on dabrafenib/trametinib and to add palbociclib based on additional alteration in CDKN2A. Two months after the addition of palbociclib, restaging scan with PET/CT scan showed resolution of 18F-fluorodeoxyglucose–avid lung nodules as well as improvement in rectal lesion (middle to right). PFS was 11.3 months without significant toxicities.