KRAS Wt Tumors Research Articles

Adjuvant Hepatic Artery Infusion Chemotherapy is Associated With Improved Survival Regardless of KRAS Mutation Status in Patients With Resected Colorectal Liver Metastases: A Retrospective Analysis of 674 Patients.

To investigate the impact of adjuvant hepatic artery infusion (HAI) in relation to KRAS mutational status in patients with resected colorectal cancer liver metastases (CRLM). Patients with KRAS-mutated CRLM have worse outcomes after resection. Adjuvant HAI chemotherapy improves overall survival after liver resection. Patients with resected CRLM treated at MSKCC with and without adjuvant HAI who had available KRAS status (wild-type, WT; mutated, MUT) were reviewed from a prospectively maintained institutional database. Correlations between KRAS status, adjuvant HAI, clinical factors, and outcomes were analyzed. Cox proportional hazard model was used to adjust for confounders. Between 1993 and 2012, 674 patients (418 KRAS-WT, 256 MUT) with a median follow up of 6.5 years after resection were evaluated. Fifty-four percent received adjuvant HAI. Tumor characteristics (synchronous disease, number of lesions, clinical-risk score, 2-stage hepatectomy) were significantly worse in the HAI group; however, there were more patients with resected extrahepatic metastases in the no-HAI group. In KRAS-WT tumors, 5-year survival was 78% for patients treated with HAI versus 57% for patients without HAI [hazard ratio (HR) 0.51, P < 0.001]. In KRAS-MUT tumors, 5-year survival was 59% for patients treated with HAI versus 40% for patients without HAI (HR 0.56, P < 0.001). On multivariate analysis, HAI remained associated with improved OS (HR 0.53, P < 0.002) independent of KRAS status and other clinicopathologic factors. Adjuvant HAI after resection of CRLM is independently associated with improved outcomes regardless of KRAS mutational status. Adjuvant HAI may mitigate the worse outcomes seen in patients with resectable KRAS-MUT CRLM.

Integrative molecular profiling and response to chemotherapy on the COMPASS trial.

188 Background: Predictive mutational and transcriptional features in advanced PDAC are needed for improved patient stratification and treatment selection. Methods: Patients (pts) on the COMPASS trial with advanced PDAC are prospectively recruited prior to first-line chemotherapy for WGS and RNAseq. Tumor tissue undergoes enrichment by laser capture microdissection with genomic analyses available within eight wks. Tumor responses and clinical outcomes in this maturing cohort were correlated with molecular characteristics. Results: 157 pts underwent a tumor biopsy between Dec 2015 and Jul 2018 with over 95% success in achieving results. 141 genomes have been reported in pts planned to receive chemotherapy; the median time from biopsy to report was 35.5 days. In the ITT population,118 pts are summarised for first-line response. The median age was 63 yrs (29-81), 55% were male, and 16% had locally advanced disease. 66 (56%) received modified FFX as first-line treatment. 25 (21%) tumors displayed the Moffitt basal-like RNA signature which associated with chemotherapy resistance, with tumor shrinkage mainly observed in the classical RNA subtype (p = 0.002). GATA6 expression (log10 scale) clearly separated Moffitt subgroups with classical tumors exhibiting high expression (p &lt; 0.0001). GATA6 in situ hybridization strongly correlated with RNAseq, (r = 0.89, p &lt; 0.001) with strong correlation also seen with GATA6 IHC. The median overall survival (OS) in the classical group was 8.5 mths vs. 6.6 mths in the basal-like group (HR 0.53, 95% CI 0.32-0.89 p = 0.015). In those treated with mFFX, median OS was 10.1mths in classical versus 6.6mths in the basal-like subtypes, (HR 0.32, 95% CI 0.16-0.63, p=0.001). 30% of pts had potentially actionable genetic alterations including BRAF variants (n = 4) and an NTRK3-EML4 fusion in KRAS WT tumors (8%). Signatures of homologous recombination deficiency (HRD) were found in seven pts; two additional pts with BRCA germline variants did not exhibit somatic HRD hallmarks. Conclusions: Subsets of pts with advanced PDAC have actionable variants including those with HRD signatures and patients with KRAS WT tumors. GATA6 is a putative predictive biomarker of transcriptomic subgroups which should be incorporated in clinical trials. Clinical trial information: NCT02750657.

NRG1 Fusions in KRAS Wild-Type Pancreatic Cancer.

We used whole-genome and transcriptome sequencing to identify clinically actionable genomic alterations in young adults with pancreatic ductal adenocarcinoma (PDAC). Molecular characterization of 17 patients with PDAC enrolled in a precision oncology program revealed gene fusions amenable to pharmacologic inhibition by small-molecule tyrosine kinase inhibitors in all patients with KRAS wild-type (KRASWT) tumors (4 of 17). These alterations included recurrent NRG1 rearrangements predicted to drive PDAC development through aberrant ERBB receptor-mediated signaling, and pharmacologic ERBB inhibition resulted in clinical improvement and remission of liver metastases in 2 patients with NRG1-rearranged tumors that had proved resistant to standard treatment. Our findings demonstrate that systematic screening of KRASWT tumors for oncogenic fusion genes will substantially improve the therapeutic prospects for a sizeable fraction of patients with PDAC.Significance: Advanced PDAC is a malignancy with few treatment options that lacks molecular mechanism-based therapies. Our study uncovers recurrent gene rearrangements such as NRG1 fusions as disease-driving events in KRASwt tumors, thereby providing novel insights into oncogenic signaling and new therapeutic options in this entity. Cancer Discov; 8(9); 1087-95. ©2018 AACR.This article is highlighted in the In This Issue feature, p. 1047.

Prognostic significance of CEACAM5mRNA-positive circulating tumor cells in patients with metastatic colorectal cancer.

Τo evaluate the clinical relevance of CEACAM5mRNA-positive circulating tumor cells (CTCs) in patients with metastatic colorectal cancer (mCRC). Peripheral blood was obtained from 436 patients with mCRC before the initiation of systemic therapy. A second sample was obtained on treatment assessment from 296 (67.9%) patients. The detection of CEACAM5mRNA-positive CTCs was performed using a real-time PCR assay. The patients' median age was 67years and PS (EGOG 0-1) 92%; KRAS exon 2 and BRAFV600E mutated primary tumors were identified in 31.9% and 6.4% of the tested patients, respectively, whereas metastasectomy was performed in 17.7% of the patients. Circulating CEACAM5mRNA-positive CTCs were detected in 125 (28.7%) and 85 (28.7%) patients at baseline and on treatment assessment, respectively. The detection of CEACAM5mRNA-positive cells was revealed, in multivariate analysis, as an independent prognostic factor associated with decreased PFS (HR 1.6; 95% CI 1.1-2.5; p = 0.026) and OS (HR 2.2; 95% CI 1.3-3.2; p < 0.001). The detection of CEACAM5mRNA-positive CTCs in patients with KRAS and BRAFV600E mutations was correlated with shorter PFS (p = 0.041 and p = 0.022, respectively). Moreover, OS was significantly shorter in patients with CEACAM5+/KRAS mutations compared to those with CEACAM5+/KRAS wt tumors (p = 0.023). Detection of peripheral blood CEACAM5mRNA-positive CTCs is an adverse prognostic factor correlated with poor clinical outcome in patients with mCRC, especially in patients with KRAS and BRAF mutated tumors.

Impact of second-line cetuximab-containing therapy in patients with KRAS wild-type metastatic colorectal cancer: results from the ITACa randomized clinical trial

The ITACa trial was designed to define the role of cetuximab (Cet) and bevacizumab (Bev) in combination with standard chemotherapy (CT, FOLFIRI or FOLFOX4) as first- and second-line treatment in metastatic colorectal cancer. All patients with WT KRAS tumors who had been enrolled in the first-line trial were randomized onto two independent second-line trials: CT or CT + Cet (study 2A) and CT + Bev or CT + Bev + Cet (study 2B). Patients with mutated KRAS were not eligible for randomization and were treated with CT alone (study 2A) or CT + Bev (study 2B). The primary endpoint was progression-free survival (PFS). 48 and 56 KRAS WT patients were randomized while 31 and 40 KRAS mutated patients were treated without randomization. Study 2A: median PFS was 3.4 (95%CI 2.3–4.6) and 6.2 (95%CI 4.3–7.8) months for the CT and CT + Cet arms, respectively, with a hazard ratio (HR) = 0.64 (95%CI 0.35–1.16, p = 0.144). Study 2B: median PFS was 7.7 (95%CI 4.1–10.1) and 4.9 (95%CI 3.2–7.0) months for CT + Bev and CT + Cet + Bev arms, respectively, with a HR = 1.31 (95%CI 0.76–2.26, p = 0.330). Notwithstanding limitations due to the small sample size, among patients with WT KRAS the addition of Cet to second-line CT increased PFS, whereas the addition of Cet to CT + Bev was associated with worse PFS.

Abstract 4856: High mRNA expression of splice variant SYK short correlates with poor hepatic metastasis free survival in untreated lymph node negative colon cancer patients

Abstract Introduction: In lymph node negative (LNN) colon cancer 20% of the patients will develop recurrence of disease. Identification of these patients is an unmet need. SYK, a protein kinase, has been ascribed both a tumor promoter and suppressor role in epithelial cancers. The prognostic value of SYK and its splice variants, largely uknown in colorectal cancer however, was explored in a clinically well-defined cohort of colon cancer patients. Methods: Total mRNA expression of SYK [SYK(T)] and of its two splice variants SYK short(S) and SYK long(L) were measured using RT-qPCR in a clinically well-defined prospectively collected cohort of 240 colon cancer patients (n=160 untreated lymph node negative [LNN] and n=80 adjuvant treated lymph node positive [LNP] patients) selected from the MATCH-cohort. mRNA expression levels were related to microsatellite instability (MSI), mRNA expression of epithelial (EPCAM), stromal (BGN, FAP, INHBA) and infiltrate markers (VEGFA, CD45), known CRC mutations (n=238), and disease free (DFS), hepatic metastasis free (HFS) and overall survival (OS). Results: Overall increased SYK levels were associated with stage I/II, a left-sided located primary tumor and MicroSatellite Stability (MSS). However, these associations and their interrelation differed significantly between SYK(T), SYK(S) and SYK(L) expression implicating an added value for measuring mRNA expression of the splice variants next to SYK(T). SYK(T), SYK(S) and SYK(L) levels all showed a significant positive correlation with the expression of EPCAM, FAP was weakly negatively associated with SYK(S) and VEGFA was weakly positively correlated with SYK(T) and SYK(S). This suggests a higher expression of SYK in epithelial-rich, stromal-poor tumors. SYK(T) and SYK(S) expression was significantly lower in tumors with a BRAF or PTEN mutation (mt) compared to wild type (wt) tumors. Although others reported differential expression of SYK between KRAS-dependent and KRAS-independent cell lines and KRAS mt versus wt tumors in 221 TCGA-samples (p=0.008), we observed no significant differences for expression of SYK(T), SYK(S) and SYK(L) between KRAS-mutant (mt) and KRAS-wild type (wt) tumors. In the LNN group, using univariate Cox regression analysis increasing mRNA expression of SYK(T) (HR=2.05 95%CI=1.01-4.17 p=0.047) and SYK(S) (HR=1.83 95%CI=1.09-3.05 p=0.021) was associated with worse HFS, which remained significant for SYK(S) when correcting for the number of assessed lymph nodes (HR=1.83; 95% CI=1.08-3.12; p=0.026 and HR=1.27; 95%CI=1.009-1.60; p=0.042). No other significant associations between SYK(T), SYK(S) and SYK(L), and DFS, HFS and OS were observed. Conclusion: In our untreated LNN colon cancer cohort SYK(S) is a pure prognostic marker for HFS. These results may help to identify LNN patients at overall low risk to develop liver metastases. Note: This abstract was not presented at the meeting. Citation Format: Robert R. Coebergh van den Braak, Anieta M. Sieuwerts, Zarina S. Lalmahomed, Sandra Bril, Annemieke M. Timmermans, Vanja de Weerd, Michelle van der Vlugt - Daane, Anne van Galen, Shanshan Xiang, Katharina Biermann, John A. Foekens, John W. Martens, Jan N. IJzermans. High mRNA expression of splice variant SYK short correlates with poor hepatic metastasis free survival in untreated lymph node negative colon cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4856. doi:10.1158/1538-7445.AM2017-4856

The YAP1/SIX2 axis is required for DDX3-mediated tumor aggressiveness and cetuximab resistance in KRAS-wild-type colorectal cancer.

The mechanism underlying tumor aggressiveness and cetuximab (CTX) resistance in KRAS-wild-type (KRAS -WT) colorectal cancer remains obscure. We here provide evidence that DDX3 promoted soft agar growth and invasiveness of KRAS-WT cells, as already confirmed in KRAS-mutated cells. Mechanistically, increased KRAS expression induced ROS production, which elevated HIF-1α and YAP1 expression. Increased HIF-1α persistently promoted DDX3 expression via a KRAS/ROS/HIF-1α feedback loop. DDX3-mediated aggressiveness and CTX resistance were regulated by the YAP1/SIX2 axis in KRAS-WT cells and further confirmed in animal models. Kaplan-Meier and Cox regression analysis indicated that DDX3, KRAS, and YAP1 expression had prognostic value for OS and RFS in KRAS-WT and KRAS-mutated tumors, but SIX2 and YAP1/SIX2 were prognostic value only in KRAS-WT patients. The observation from patients seemed to support the mechanistic action of cell and animal models. We therefore suggest that combining YAP1 inhibitors with CTX may therefore suppress DDX3-mediated tumor aggressiveness and enhance CTX sensitivity in KRAS-WT colorectal cancer.

KRAS Mutation Status Dictates Optimal Surgical Margin Width in Patients Undergoing Resection of Colorectal Liver Metastases.

The optimal tumor-free margin width remains controversial and may be inappropriate to investigate without considering differences in the underlying tumor biology. R1 resection was defined as margin clearance less than 1mm. R0 resection was further divided into 3 groups: 1-4, 5-9, and ≥10mm. The impact of margin width on overall survival (OS) relative to KRAS status [wild type (wtKRAS) vs. mutated (mutKRAS)] was assessed. A total of 411 patients met inclusion criteria. Median patient age was 58years (interquartile range, 49.7-66.7); most patients were male (n=250; 60.8%). With a median follow-up of 28.3months, median and 5-year OS were 69.8months and 55.1%. Among patients with wtKRAS tumors, although margin clearance of 1-4mm or more was associated with improved OS compared to R1 (all P<0.05), no difference in OS was observed when comparing margin clearance of 1-4mm to the 5-9mm and the ≥10mm groups (all P>0.05). In contrast, among patients with mutKRAS tumors, all three groups of margin clearance (1-4, 5-9, and ≥10mm) fared no better in terms of 5-year survival compared to R1 resection (all P>0.05). While a 1-4mm margin clearance in patients with wtKRAS tumors was associated with improved survival, wider resection width did not confer an additional survival benefit. In contrast, margin status-including a 1cm margin-did not improve survival among patients with mutKRAS tumors.

Tumor Biology Rather Than Surgical Technique Dictates Prognosis in Colorectal Cancer Liver Metastases

IntroductionThe interplay of tumor biology and surgical margin status after resection for colorectal liver metastasis (CRLM) remains controversial. Consequently, we sought to determine the impact of surgical margin status on overall survival (OS) stratified by KRAS mutational status. Materials and MethodsFour hundred eighty-five patients with known KRAS mutational status were identified. Clinicopathologic and long-term survival data were collected and assessed. ResultsOn pathology, most patients (n = 380; 78.3 %) had an R0 margin, while 105 (21.7 %) had an R1. Roughly two thirds of tumors were KRAS wild type (wtKRAS) (n = 307, 63.3 %), while 36.7 % (n = 178) had KRAS mutations (mutKRAS). Median and 5-year OS of the entire cohort was 65.8 months and 53.8 %, respectively. An R1 resection was associated with worse 5-year OS compared with R0 (42.4 % vs. 57.1 %; hazard ratio (HR) 1.82, 95 % CI 1.28–2.57; P = 0.001). After controlling for KRAS status, the survival benefit associated with an R0 resection persisted only among patients with wtKRAS tumors (HR 2.16, 95 % CI 1.42–3.30; P < 0.001). In contrast, surgical margin had no impact on OS among patients with mutKRAS tumors (5-year OS R0, 40.7 % vs. R1, 46.7 %; HR 1.34, 95 % CI 0.73–2.48; P = 0.348). ConclusionThe impact of margin status differed by KRAS mutation status. An R0 margin only provided a survival benefit to patients with wtKRAS tumors. Tumor biology and not surgical technique determined prognosis.

FOLFIRI plus cetuximab in patients with liver-limited or non-liver-limited RAS wild-type metastatic colorectal cancer: A retrospective subgroup analysis of the CRYSTAL study

BackgroundAdding cetuximab to first-line FOLFIRI in the phase 3 CRYSTAL trial significantly improved progression-free survival (PFS), overall survival (OS), and objective response rate (ORR) in patients with KRAS wild-type (wt) or RAS wt metastatic colorectal cancer (mCRC). In this retrospective subgroup analysis of CRYSTAL, we investigated benefit of treatment in patients with KRAS wt or RAS wt tumors according to whether patients had liver-limited disease (LLD) or non-LLD, including assessing the role of cetuximab in downsizing metastases and conversion rates from initially unresectable to resectable disease. MethodsPFS, OS, ORR, and R0 resection rates were analyzed according to treatment arm for the LLD and non-LLD subgroups. ResultsOf the 367 patients with RAS wt tumors, 89 (24%) had LLD and 278 (76%) had non-LLD. Within the RAS wt LLD and non-LLD subpopulations, demographic and baseline characteristics were comparable between treatment arms. In patients with RAS wt LLD, adding cetuximab to FOLFIRI significantly improved PFS (hazard ratio [HR][95% CI] = 0.21[0.09–0.49]) and ORR (odds ratio [OR][95% CI] = 8.99[3.17–25.52]), and numerically improved OS (HR[95% CI] = 0.65[0.38–1.10]) and R0 resection rate (OR[95% CI] = 2.68[0.63–11.43]) relative to FOLFIRI alone. In patients with RAS wt non-LLD, adding cetuximab to FOLFIRI significantly improved PFS (HR[95% CI] = 0.65[0.46–0.93]), OS (HR[95% CI] = 0.71[0.54–0.93]), ORR (OR[95% CI] = 2.44[1.49–3.98]), and—numerically—R0 resection rate (OR[95% CI] = 5.94[0.79–44.88]). Similar results were obtained from the KRAS wt population. ConclusionsAdding cetuximab to first-line FOLFIRI appears to improve clinical outcomes and R0 resection rates in KRAS wt and RAS wt mCRC patients with LLD as well as in those with non-LLD.

Patterns in KRAS testing and EGFRi therapy across lines of treatment for metastatic colorectal cancer in Canada: A retrospective analysis.

665 Background: Selection and sequencing of treatment regimens for individual metastatic colorectal cancer (mCRC) patients is governed by the goals of maintaining reasonable quality of life while extending survival. The timing of KRAS testing and its effect on EGFRi therapy is poorly described. The goals of this analysis wereto describe rates and timing of KRAS testing relative to EGFRi therapy for Canadian patients diagnosed with mCRC. Methods: A retrospective chart review conducted at 6 Canadian centres included patients diagnosed with mCRC from Jan 1, 2009 onwards, who commenced 1st-line systemic treatment for mCRC between Jan 1–Dec 31, 2009. Information on the proportion of patients who received 2nd, 3rd, or subsequent lines of systemic therapy for mCRC was determined and the rates and timing of KRAS testing was ascertained. Results: 200 patients commenced 1st-line therapy and the median age was 62 yr; 78% had mCRC at the time of diagnosis. The proportions of patients who started 2nd, 3rd, and 4th lines of systemic therapy were 70%, 30%, and 15%, respectively. 103 (52%) patients had KRAS testing; 6%, 18%, 57%, 16%, 2%, and 1% of patients were tested at diagnosis or before the 2nd, 3rd, 4th, 5th and 6th lines of therapy, respectively. Median time from testing to EGFRi treatment was 105 (range, 7–1192) days, and varied by site. The frequency of KRAS testing for patients ranged from 30%–70% across study sites; across provinces, the frequency of testing ranged from 46%–60%. 38/68 (56%) of patients with wt KRAS tumors received EGFRi; 31 (46%) patients received EGFRi therapy as next therapy following KRAS testing. 19 (28%) died and 4 were lost to follow-up within 120 days of KRAS testing with no other therapy. 2 additional patients with unknown KRAS status received EGFRi (1 without KRAS testing; 1 undetermined). Conclusions: KRAS testing occurred after starting 2nd line in 76% of cases and varied by site and province. About half of patients underwent KRAS testing and 56% of those patients with wt KRAS tumors received an EGFRi. The short time interval between (K)RAS testing and EGFRi therapy may point to the need for earlier testing if EGFRi therapy is to be used in earlier lines of therapy.

Abstract 4842: Pharmacodynamic stratification of metastatic colorectal cancer patients using genomic datasets

Abstract Previously, the mutation status of KRAS was the only validated predictive biomarker for metastatic colorectal cancer (CRC). While KRAS mutated tumors demonstrated resistance to epidermal growth factor (EGFR) inhibitors like cetuximab, KRAS WT and EGFR-expressing tumors were predicted to be responsive. However, KRAS WT metastatic colorectal cancer (CRC) patients have a poor prognosis even with EGFR inhibitor therapy as not all KRAS WT CRCs are responsive to such targeted agents. A gene expression based RAS signature score was developed based on multiple tumor tissue samples to identify RAS activated tumors independent of mutations in the KRAS gene [1, 2]. To further refine this score and define technologies that can be used on FFPE samples isolated in a clinical setting, we analyzed DNA and RNA derived from fifty-five (55) FFPE preserved colorectal cancer tumor biopsies using multiple sequencing, digital and array-based technologies. These samples were selected from a CRC cohort in which the initial gene expression-based RAS signature score was calculated utilizing data compiled from fresh frozen (FF) tumor samples from the same 55 patients. The 55 samples were selected for this study as they had representative samples with high, medium and low RAS signature scores. Transcriptomic analyses (RNA-Seq, Affymetrix microarrays, Nanostring and Targeted RNA-Seq) were performed on all 55 FFPE samples and three new RAS scores were calculated from the gene expression datasets. These RAS scores were based on different gene signatures (1) an 18 gene signature (2) a 13 gene signature, and (3) a 147 gene signature. A significant correlation was identified between RAS scores calculated from the 18 and 13 gene signatures (Correlation coefficient ∼ 0.88 and ∼0.76 respectively, p-value &amp;lt; 0.0001). To further refine gene expression signatures, samples were grouped based upon their mutation status obtained by whole exome sequencing (WES) and targeted DNA sequencing data (Illumina TruSight and LifeTech Cancer Panels). In our sample set, the 18 gene RAS score was found to be dependent on the mutation status of KRAS. Further analysis is being carried out to better understand the relationship between the calculated RAS signature scores and the mutation status of other genes. This analysis will lead to the development of a novel genomic signature for better pharmacodynamic stratification of colorectal carcinoma patients. 1. Loboda A et al. A gene expression signature of RAS pathway dependence predicts response to PI3K and RAS pathway inhibitors and expands the population of RAS pathway activated tumors. 2. BMC Medical Genomics 2010, 3:26Dry JR et al. Transcriptional Pathway Signatures Predict MEK Addiction and Response to Selumetinib (AZD6244). Cancer Res. 2010 Mar 15;70(6):2264-73. Citation Format: Sharon Austin, Fang Yin Lo, Kellie Howard, Mollie McWhorter, Heather Collins, Amanda Leonti, Lindsey Maassel, Christopher Subia, Tuuli Saloranta, Nicole Heying, Leila Ritter, Kerry Deutsch, James Cox, Steven Anderson, Anup Madan, Timothy Yeatman. Pharmacodynamic stratification of metastatic colorectal cancer patients using genomic datasets. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4842. doi:10.1158/1538-7445.AM2015-4842

Gene expression markers of efficacy and resistance to cetuximab treatment in metastatic colorectal cancer: results from CALGB 80203 (Alliance).

Formalin-fixed, paraffin-embedded tumor samples from CALGB 80203 were analyzed for expression of EGFR axis-related genes to identify prognostic or predictive biomarkers for cetuximab treatment. Patients (238 total) with first-line metastatic colorectal cancer (mCRC) were randomized to FOLFOX or FOLFIRI chemotherapy ± cetuximab. qRT-PCR analyses were conducted on tissues from 103 patients at baseline to measure gene expression levels of HER-related genes, including amphiregulin (AREG), betacellulin (BTC), NT5E (CD73), DUSP4, EGF, EGFR, epigen (EPGN), epiregulin (EREG), HBEGF, ERBB2 (HER2), ERBB3 (HER3), ERBB4 (HER4), PHLDA1, and TGFA. The interactions between expression levels and treatment with respect to progression-free survival (PFS) and overall survival (OS) were modeled using multiplicative Cox proportional hazards models. High tumor mRNA levels of HER2 [hazard ratio (HR), 0.64; P = 0.002] and EREG (HR, 0.89; P = 0.016) were prognostic markers associated with longer PFS across all patients. HER3 and CD73 expression levels were identified as potential predictive markers of benefit from cetuximab. In KRAS wild-type (WT) tumors, low HER3 expression was associated with longer OS from cetuximab treatment, whereas high HER3 expression was associated with shorter OS from cetuximab treatment (chemo + cetuximab: HR, 1.15; chemo-only: HR, 0.48; Pinteraction = 0.029). High CD73 expression was associated with longer PFS from cetuximab treatment in patients with KRAS-WT (chemo + cetuximab: HR, 0.91; chemo-only: HR, 1.57; Pinteraction = 0.026) and KRAS-mutant (Mut) tumors (chemo + cetuximab: HR, 0.80; chemo-only: HR, 1.29; P = 0.025). Gene expression of HER3 and CD73 was identified as a potential predictive marker for cetuximab. These data implicate HER axis signaling and immune modulation as potential mechanisms of cetuximab action and sensitivity.

Association of EGFR CA simple sequence repeat 1 (CA-SSR1) variant with cetuximab (cet)-induced skin toxicity (ST) in Japanese metastatic colorectal cancer (mCRC) patients (pts) with overexpressed EGFR and KRAS exon 2 wild-type (KRAS wt) (JACCRO CC-05/06 AR).

582 Background: Associations of EGFR gene copy number (GCN) and EGFR CA-SSR1 polymorphism with clinical outcome still remain controversial in mCRC pts treated with cet. Interethnic differences in the repeat number of EGFRCA-SSR1 have been reported between Caucasian and Asians, however, there have been few reports about the associations with clinical outcome in Japanese pts. The aim of this study was to evaluate the predictive value of two biomarkers in Japanese mCRC pts treated with cet. Methods: This study enrolled 77 pts with tumor available from two prospective clinical trials evaluating combination of cet with oxaliplatin-based chemotherapy as first-line treatment in mCRC pts with KRAS wt and EGFR-expressing tumors, modified FOLFOX6 (n=28/57, UMIN000004197) and SOX (n=49/67, UMIN000007022). Genomic DNA isolated by macro-dissection from tissue was screened for the EGFR GCN determined by FISH on FFPE tumor specimens with the cut-off value of 2.9, adopted as previously reported (Cappuzzo F, et al., 2008), and the EGFR CA-SSR1 short (S; ≤19) / long (L; ≥20) determined by PCR amplification and fragment length analysis using capillary electrophoresis. Associations of the biomarkers with efficacy, including response, progression-free survival, and overall survival, were evaluated. Additional analysis was addressed at a possible association between the EGFRCA-SSR1 and ST. Results: The frequency of the L alleles of the EGFR CA-SSR1 was 64%. There was no significant association between two biomarkers and efficacy. However, the EGFR CA-SSR1 variant significantly correlated with ST evaluated on 8 weeks after initial administration. The rate of ST with grade 2 or 3 was 33% (10/30), 19% (5/26), and 64% (7/11) in LL, SL, and SS genotype, respectively (fisher's exact test p=0.03). Conclusions: Our study provides first evidence that the EGFR CA-SSR1 variant was associated with cet-induced ST in Japanese mCRC pts. No predictive value for cet treatment could be identified in our screened biomarkers, however, extended RAS mutations analysis is warranted (UMIN000010635).

EGFR ligands and ERCC1 mRNA expression to predict clinical outcome in Japanese (JPN) patients (pts) with metastatic colorectal cancer (mCRC) harboring overexpressed EGFR and KRAS exon 2 wild-type (KRAS wt) treated with cetuximab (cet) plus oxaliplatin-based chemotherapy (JACCRO CC-05/06 AR).

618 Background: Previous studies have reported that EGFR ligands, amphiregulin (AREG) and epiregulin (EREG), are potential predictive markers for cet and high ERCC1 expression is associated with resistance to platinum-containing chemotherapy. However, there are still few studies to assess predictive values of these expression levels in JPN mCRC pts. The aim of this study was to evaluate the values in JPN mCRC pts treated with cet. Methods: This study enrolled 77 pts with tissue available from 2 prospective clinical trials evaluating combination of cet with oxaliplatin-based chemotherapy as first-line treatment in pts with KRAS wt and EGFR-expressing tumors, modified FOLFOX6 (n=28/57, UMIN000004197) and SOX (n=49/67, UMIN000007022). Total RNA isolated by macro-dissection from tissue was screened by RT-PCR for mRNA expression levels of AREG, EREG, and ERCC1, the cut-off values of which were based on the optimal cut-off value using the maximal chi-square approach on tumor response with adjusted p values. The same cut-off values were used to evaluate associations with progression-free survival (PFS) and overall survival (OS). Results: Median age and follow-up time were 63 (range: 39-79) years old and 24.7 (range: 5.5-39.3) months, respectively. In 73 evaluable pts, response rate (RR) was 77 % (95% CI, 67 to 86), but no significant association was seen in RR. Pts with high AREG expression (&gt;1.59) had significantly longer PFS (11.6 vs. 6.6 months, HR: 0.53, logrank p=0.047) but not OS in 67 assessable pts. Pts with low ERCC1 expression (≤1.35) had significantly longer OS (42.8 vs. 22.0 months, HR: 2.46, logrank p=0.014) but a trend toward longer PFS (11.6 vs. 8.9 months, HR: 1.63, logrank p=0.092) in 68 assessable pts. On the other hand, high EREG expression (&gt;3.21) was not associated with clinical outcome in 64 assessable pts. Conclusions: Our study identified predictive roles for AREG and ERCC1 mRNA expression in JPN mCRC pts treated with cet plus oxaliplatin. However, extended RAS mutations analysis is warranted (UMIN000010635).

Genomic markers of panitumumab resistance including ERBB2/HER2 in a phase II study of KRAS wild-type (wt) metastatic colorectal cancer (mCRC).

646 Background: Epidermal Growth Factor Receptor Inhibitors (EGFRi) are effective agents in the treatment of mCRC but up to one third of KRAS wt tumors do not respond. The primary objective was to identify a gene expression signature associated with primary resistance to panitumumab (Pmab) monotherapy and identify genomic markers associated with resistance. Methods: Patients with previously treated codon 12/13 KRAS wt mCRC were administered Pmab 6mg/kg IV q2weeks in a prospective phase II study. The primary endpoint was tumor response by RECIST and patients with progressive disease 8 weeks after initial Pmab were classified as resistant. Gene expression analysis was performed using a Nanostring-based assay with candidate genes implied in EGFRi resistance expression patterns. Differentially expressed genes were identified by Significance Analysis of Microarrays. Further interrogation of the HER2 signaling pathway was performed by immunohistochemistry and in situ hybridization. Illumina-based targeted resequencing was conducted on 20 cancer-implicated genes. Results: Of 34 patients treated with Pmab, 11 (32%) had progressive disease at 8 weeks and were classified Pmab resistant. A 5-gene expression signature was significantly associated with Pmab resistance (p = 0.001), including ERBB2, MLPH, IRX3, c11orf9, and KLK6. HER2 protein was overexpressed in 4/11 resistant and 0/21 responding cases (p = 0.035). Two (2) resistant tumors had ERBB2 gene amplification only, and one demonstrated both ERBB2 amplification and mutation. A non-codon 12/13 RAS mutation occurred in one treatment-resistant patient and was mutually exclusive with ERBB2 mutation, amplification, or HER2 overexpression. Four BRAF mutations were detected, two occurring in resistant patients. Conclusions: In this prospective study, a 5-gene signature including ERBB2 was associated with primary resistance to single agent Pmab. A subgroup of patients had evidence of aberrant HER2/ERBB2 signaling by protein over-expression and gene amplification. Results suggest a significant portion of KRAS wt tumors unresponsive to EGFRi may be identified by gene signature analysis. Clinical trial information: NCT00853931.

Abstract 2835: Telomere length: A novel biomarker for anti-EGFR therapy in colorectal cancer

Abstract Background: Telomeres (T) consist of thousands of copies of TTAGGG tandem repeats capping the ends of the chromosome. Studies have linked short T length (TL) with premalignant and malignant stages of colorectal cancer (CRC) carcinogenesis. However, the relation of TL to response to drugs used in treatment of CRC has not been conclusively determined. In this study, we used a panel of CRC cell lines and analyzed each TL, and determined relationship to drug sensitivity. We validated our in vitro findings in a pool of samples collected from patients with metastatic CRC (mCRC) undergone anti-EGFR therapy at our institution. Methods: DNA was isolated from CRC cell lines grown in culture, and from formalin fixed paraffin embedded specimens. Cells were cultured and treated with varying concentrations of cetuximab, oxaliplatin, 5-FU, camptothecin, aspirin, sulindac, trichostatin, curcumin, and butyrate. TL was measured using a qRT-PCR method that provided the relative T copy number, compared to reference single copy number of the housekeeping gene, β-Globin (S), which resulted in a T/S ratio. One T/S ratio unit is equivalent to a mean TL of 4,270 base pairs. Results: In a panel of 22 CRC cell lines, TL was independently predicative of cetuximab sensitivity. When split at the median, thee cell lines with a shorter TL had a mean cetuximab sensitivity of 18.6% (growth inhibition at 20 ug/ml at 72 hours by MTT assay), as compared to 39.7% in those cells with longer TL; p=0.02. Response to all of the other drugs did not bear any relationship to TL. To further validate our in vitro finding, we analyzed the TL in patients with mCRC who had been treated with cetuximab or panitumumab. In this sample of 75 patients, 118 tissue samples were identified (66 primary, 52 metastases, 29 matched pairs). Forty five patients had received greater than 4 weeks of therapy and were considered for clinical correlates. Patients with K-Ras WT tumors had a superior progression free survival (PFS) and overall survival than those with K-Ras mutated tumors (p&amp;lt;0.05). Importantly, among the patients with K-Ras WT tumors, those with a longer TL had a superior outcome with median PFS of 28.1 weeks as compared to those with shorter TL, median14.7 weeks, p=0.019. Conclusion: TL is a potential biomarker predictive of sensitivity to cetuximab, in vitro, and of outcome for patients treated with cetuximab/panitumumab, as measured by PFS. These findings will be validated in a larger dataset, and the mechanistic basis for this finding is currently under investigation. Citation Format: Titto A. Augustine, Mahadi A. Baig, John M. Mariadason, Radhashree Maitra, Sanjay Goel. Telomere length: A novel biomarker for anti-EGFR therapy in colorectal cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2835. doi:10.1158/1538-7445.AM2014-2835

Calgb/Swog 80405: Phase III Trial of Irinotecan/5-Fu/Leucovorin (Folfiri) or Oxaliplatin/5-Fu/Leucovorin (Mfolfox6) with Bevacizumab (Bv) or Cetuximab (Cet) for Patients (Pts) with Expanded Ras Analyses Untreated Metastatic Adenocarcinoma of the Colon Or Rectum (Mcrc)

ABSTRACT Aim: FOLFIRI or mFOLFOX6, combined with BV or CET, are 1st-line treatments for MCRC. The optimal antibody combination is unknown. Methods: Pts with RAS wt (codons 12 and 13) MCRC and performance status 0-1 received FOLFIRI or mFOLFOX6 (MD/pt choice at enrollment) and randomized to either CET 400 mg/m2 X 1, then 250 mg/m2 qw or BV 5 mg/kg q2w. The original study included unselected MCRC pts receiving FOLFIRI or mFOLFOX6 and randomized to CET, BV or both. After 1420 pts accrued the study amended as follows: only pts w/ KRAS wt tumors (codon 12 and 13) were included. Accrual goal was 1142 pts Expanded RAS was tested in all wt ras exon 2 using beaming technology including KRAS exon 3,4 and NRAS exon 2, 3 and 4 with a detection sensitivity of 0.01%. Subsequent Rx not mandated. 1° endpoint was overall survival (OS). Results: Between 11/2005 and 3/2012, 3058 unselected pts enrolled, 2334 KRAS wt pts randomized; final N =1137 (333 pre-amend eligible retrospective KRAS test, 804 post-amend), median f/u = xx mos; Median age–59 y; 61% male. Chemo/BV–559; chemo/CET–578. FOLFIRI = 26.6%, mFOLFOX6 = 73.4%. From xx patients tumor samples were available for expanded ras analyses. We identified xx% mutations in kras exon 3 and 4 and NRAS in exon 2,3 and 4 in patients with wt KRAS exon2. The OS in patients with wt RAS treated with chemo/BV v. chemo/CET = xx (xx - xx) v. xx (xx - x) mos; HR = xx (xx, xx) (p value = xxx). PFS (by investigator): chemo/BV v. chemo/CET: xxx (xx - xx) v. xxx (xx - xx) mos. On-study toxicity and deaths as expected. Conclusions: Updated analysis will be shown at meeting. Disclosure: H. Lenz: Ad Board Genentech/Roche, MerckKG, EMD, BMS. All other authors have declared no conflicts of interest.

539P - Tumor Heterogeneity and Efficacy of First-Line Cetuximab + Folfiri in Kras Mutant Metastatic Colorectal Cancer (Mcrc) Patients (Pts) of the Capri Goim Trial

ABSTRACT Aim: Cetuximab treatment in mCRC is restricted to pts with KRAS and NRAS wt tumors. Recent data suggest that low levels of KRAS mutant alleles might confer resistance to anti-EGFR therapies. Methods: In the CAPRI GOIM trial, KRAS exon 2 wt mCRC pts received first line cetuximab plus FOLFIRI. Primary endpoint was progression-free survival (PFS). Tumor samples were assessed by next generation sequencing (NGS) with the Ion AmpliSeq™ Lung and Colon Cancer Panel that covers over 500 mutations in hot spot regions of 22 genes. The allelic frequency of mutant KRAS, NRAS, BRAF and PIK3CA was normalized for the neoplastic cell content and multiplied by 2 to obtain the Heterogeneity Score (HS), which corresponds to the fraction of neoplastic cells carrying a specific mutation. Results: NGS analysis was performed in 182 cases. Forty-five cases (24.7%) had mutations in KRAS, 24 (13.2%) in PIK3CA, 15 (8.2%) in BRAF, and 13 (7.1%) in NRAS. KRAS and NRAS mutations were mutually exclusive, whereas PIK3CA and BRAF mutations were detected in RAS mutant tumors. The KRAS HS ranged between 12 and 260, indicating a wide heterogeneity of KRAS mutant alleles content in CRC. However, the KRAS HS mean value of 90,93 and median value of 84,44 suggested that in most CRC the majority of neoplastic cells carry a mutant KRAS allele. Similar findings were observed for NRAS (HS range 35,5-146,67; mean 102,77; median 117,14). In contrast, in BRAF (HS range 17,14-120; mean 54,82; median 54,29) and PIK3CA (HS range 14,29-120; mean 54,82; median 54,29) mutant cases only a fraction of neoplastic cells seem to carry the mutant allele. The RR was 70% in KRAS mutant pts with an HS 40 pts (high KRAS; n.35); mPFS were 7,97 and 8,37 months for low KRAS and high KRAS subgroups, respectively. Importantly, 7/10 low KRAS tumors had additional mutations in PIK3CA (n.6), TP53 (n.4), BRAF (n.2), ERBB2 (n.1), FGFR3 (n.1), and/or FBXW7 (n.1). Conclusions: Pts carrying low levels of mutant KRAS alleles showed a mPFS similar to pts with high KRAS mutant allele frequency. However, this phenomenon could be due to the complex mutational profile of CRC rather than to the KRAS mutational status. Disclosure: N. Normanno: Advisory Boards: Merck Serono, Amgen; F. Ciardiello: Advisory Boards: Roche, Merck Serono, Astellas, Bayer. All other authors have declared no conflicts of interest.

CALGB/SWOG 80405: Phase III trial of irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (mFOLFOX6) with bevacizumab (BV) or cetuximab (CET) for patients (pts) with KRAS wild-type (wt) untreated metastatic adenocarcinoma of the colon or rectum (MCRC).

LBA3 Background: Irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/leucovorin (mFOLFOX6), combined with bevacizumab (BV) or cetuximab (CET), are first-line treatments for metastatic adenocarcinoma of the colon or rectum (MCRC). The optimal antibody combination is unknown. Methods: Patients (pts) with KRAS wild-type (wt)(codons 12 and 13) MCRC and performance status 0-1 received FOLFIRI or mFOLFOX6 (MD/pt choice at enrollment) and randomized to either CET 400 mg/m2 X 1, then 250 mg/m2 qw or BV 5 mg/kg q2w. The original study included unselected MCRC pts receiving FOLFIRI or mFOLFOX6 and randomized to CET, BV, or both. After 1,420 pts accrued the study amended as follows: only pts with KRAS wt tumors (codon 12 and 13) were included and the combination CET + BV arm was deleted. Rx continued until progression, death, unacceptable toxicity, curative surgery; treatment holidays of 4 wks permitted. Subsequent Rx not mandated. Accrual goal was 1,142 pts. One° endpoint was overall survival (OS). Results: Between November 2005 and March 2012, 3,058 unselected pts enrolled, 2,334 KRAS wt pts randomized; final N =1137 (333 pre-amend eligible retrospective KRAS test, 804 post-amend), median f/u = 24 mos; Median age – 59 y; 61% male. Chemo/BV – 559; chemo/CET – 578. FOLFIRI = 26.6%, mFOLFOX6 = 73.4%. OS analysis planned at 849 events; efficacy futility boundary crossed at 10th interim analysis on 1/29/14. OS - chemo/BV v. chemo/CET = 29.04 (25.66 - 31.21) v. 29.93 (27.56 - 31.21) mos; HR = 0.92 (0.78, 1.09) (p value = 0.34). PFS (by investigator): chemo/BV v. chemo/CET: 10.84 (9.86 - 11.4) v. 10.45 (9.66 - 11.33) mos. There were 94 pts free of disease following surgery, median f/u 40 mos (range 8.0 - 86.0). Outcomes similar by gender. On-study toxicity and deaths as expected. Analyses underway: Expanded RAS, FOLFOX v. FOLFIRI, subsequent therapies, long-term survivors, correlates. Conclusions: Chemo/CET and chemo/BV equivalent in OS in pts KRAS wt (codons 12 + 13) MCRC; either is appropriate in first line. Overall OS of 29 + mos and 8% long-term survivors confirms progress in MCRC. The preference for FOLFOX limits chemotherapy comparison. Expanded RAS and other molecular and clinical analyses may identify subsets of pts who get more or less benefit from specific regimens. Clinical trial information: NCT00265850.