Does KRAS mutational status predict chemoresistance in advanced non-small cell lung cancer (NSCLC)?

Abstract

Background

Clinical implications of KRAS mutational status in advanced non-small cell lung cancer (NSCLC) remain unclear. To clarify this point, we retrospectively explored whether KRAS mutations could impact tumor response, and disease control rate (DCR) to first-line platinum-based chemotherapy (CT) as well as progression-free survival (PFS) or overall survival (OS).

Methods

Between June 2009 and June 2012, 340 patients with advanced (stage IIIB/IV) NSCLC were reviewed in a single institution (Institut Gustave Roussy). Two hundred and one patients had a biomolecular profile and received a platinum-based first-line CT. Patients with an unknown mutational status or with actionable alterations were excluded. We retained two groups: patients with KRAS mutated tumor (MUT) and patients with wild-type KRAS/EGFR (WT). Multivariate analyses with Cox model were used. Survival curves were calculated with Kaplan–Meier method.

Results

One hundred and eight patients were included in the analysis: 39 in the MUT group and 69 in the WT group. Baseline radiological assessment demonstrated more brain (P = 0.01) and liver (P = 0.04) metastases in MUT patients. DCR was 76% for MUT vs. 91% for WT group (P = 0.03), regardless of the type of platinum-based CT (use of pemetrexed or not). Although no statistically significant differences were found, shorter PFS (4.9 vs. 6.0 months; P = 0.79) and OS (10.3 vs. 13.2 months; P = 0.40) were observed for patients with KRAS mutated tumors in univariate analysis.

Conclusions

KRAS mutant tumors had a lower DCR after the first-line platinum-based CT, but this difference did not translate in PFS or OS. The presence of KRAS mutations may confer a more aggressive disease, with greater baseline incidence of hepatic and cerebral metastases.

Introduction

Non-small cell lung cancer (NSCLC) is the most common cause of cancer-related death worldwide with more than one million deaths every year. However, over the last ten years new therapeutic options have enriched the lung anti-cancer armamentarium, and today in the metastatic setting multiple lines of therapy have become available [1], [2], [3]. Large scale studies showed that nearly all lung cancer cells exhibit inactivation of growth inhibitor pathways (TP53, RB1, p16, STK11 and CDKN2A tumor suppressors) [4] or mutations of growth regulatory genes (KRAS, EGFR, BRAF, MEK-1, HER2, MET, EML-4-ALK, KIF5B-RET, and NKX2.1). Beyond EGFR mutations and ALK rearrangements [5], [6], other “actionable” molecular mutations may be explored and used to select targeted therapies such as amplification of HER2, FGFR1 and c-MET, rearrangements of RET or ROS1, activating mutations of HER2, FGFR2, and PI3K [7], [8], [9], [10], [11], [12], [13]. Among the most common molecular alterations observed in NSCLC lie the mutations of KRAS. These mutations occur in 15–30% of NSCLC and are more frequent in adenocarcinoma (20–50%) [14]. KRAS mutations are usually mutually exclusive with EGFR mutation and ALK rearrangements [15], [16]. The prognostic value of KRAS mutations has been investigated in both the adjuvant and the metastatic setting, but its value remains controversial [17], [18]. Alongside, the predictive value of KRAS mutations to define chemosensitivity to specific chemotherapeutic agents is an area of intense debate [19].

We performed a monocentric retrospective study regarding the chemosensitivity of KRAS mutant (MUT) vs. KRAS/EGFR wild-type (WT) advanced NSCLC. We secondarily analyzed the relationship between specific KRAS mutations and patient's outcomes. The treatment response was assessed evaluating kinetics of tumor growth as well as with RECIST criteria 1.1 [22], [23].