Distinct genomic landscapes of gastroesophageal adenocarcinoma depending on PD-L1 expression identify mutations in RAS-MAPK pathway and TP53 as potential predictors of immunotherapy efficacy

BACKGROUND

The impact of molecular alterations on programmed death-ligand 1 (PD-L1) combined positive score (CPS) is not well studied in gastroesophageal adenocarcinomas (GEAs). We aimed to characterize genomic features of tumors with different CPSs in GEAs.

PATIENTS AND METHODS

Genomic alterations of 2518 GEAs were compared in three groups (PD-L1 CPS ≥ 10, high; CPS = 1-9, intermediate; CPS < 1, low) using next-generation sequencing. We assessed the impact of gene mutations on the efficacy of immune checkpoint inhibitors (ICIs) and tumor immune environment based on the Memorial Sloan Kettering Cancer Center and The Cancer Genome Atlas databases.

RESULTS

High, intermediate, and low CPSs were seen in 18%, 54% and 28% of GEAs, respectively. PD-L1 positivity was less prevalent in women and in tissues derived from metastatic sites. PD-L1 CPS was positively associated with mismatch repair deficiency/microsatellite instability-high, but independent of tumor mutation burden distribution. Tumors with mutations in KRAS, TP53, and RAS-mitogen-activated protein kinase (MAPK) pathway were associated with higher PD-L1 CPSs in the mismatch repair proficiency and microsatellite stability (pMMR&MSS) subgroup. Patients with RAS-MAPK pathway alterations had longer overall survival (OS) from ICIs compared to wildtype (WT) patients [27 versus 13 months, hazard ratio (HR) = 0.36, 95% confidence interval (CI): 0.19-0.7, P = 0.016] and a similar trend was observed in the MSS subgroup (P = 0.11). In contrast, patients with TP53 mutations had worse OS from ICIs compared to TP53-WT patients in the MSS subgroup (5 versus 21 months, HR = 2.39, 95% CI: 1.24-4.61, P = 0.016).

CONCLUSIONS

This is the largest study to investigate the distinct genomic landscapes of GEAs with different PD-L1 CPSs. Our data may provide novel insights for patient selection using mutations in TP53 and RAS-MAPK pathway and for the development of rational combination immunotherapies in GEAs.

Nintedanib for the treatment of patients with refractory metastatic colorectal cancer (LUME-Colon 1): a phase III, international, randomized, placebo-controlled study

Background

Angiogenesis is critical to colorectal cancer (CRC) growth and metastasis. Phase I/II studies have demonstrated the efficacy of nintedanib, a triple angiokinase inhibitor, in patients with metastatic CRC. This global, randomized, phase III study investigated the efficacy and safety of nintedanib in patients with refractory CRC after failure of standard therapies.

Patients and methods

Eligible patients (Eastern Cooperative Oncology Group performance status 0–1, with histologically/cytologically confirmed metastatic/locally advanced CRC adenocarcinoma unamenable to surgery and/or radiotherapy) were randomized 1 : 1 to receive nintedanib (200 mg twice daily) or placebo (twice daily), until disease progression or undue toxicity. Patients were stratified by previous regorafenib, time from onset of metastatic disease to randomization, and region. Co-primary end points were overall survival (OS) and progression-free survival (PFS) by central review. Secondary end points included objective tumor response and disease control by central review.

Results

From October 2014 to January 2016, 768 patients were randomized; 765 were treated (nintedanib n = 384; placebo n = 381). Median follow-up was 13.4 months (interquartile range 11.1–15.7). OS was not improved [median OS 6.4 months with nintedanib versus 6.0 months with placebo; hazard ratio (HR), 1.01; 95% confidence interval (CI), 0.86–1.19; P = 0.8659]. There was a significant but modest increase in PFS with nintedanib versus placebo (median PFS 1.5 versus 1.4 months, respectively; HR 0.58; 95% CI 0.49–0.69; P < 0.0001). There were no complete or partial responses. Adverse events (AEs) occurred in 97% of 384 nintedanib-treated patients and 93% of 381 placebo-treated patients. The most frequent grade ≥3 AEs were liver-related AEs (nintedanib 16%; placebo 8%) and fatigue (nintedanib 9%; placebo 6%).

Conclusions

The study failed to meet both co-primary end points. Nintedanib did not improve OS and was associated with a significant but modest increase in PFS versus placebo. Nintedanib was well tolerated.

ClinicalTrials.gov number

NCT02149108 (LUME-Colon 1).

Phase II trial of dasatinib for recurrent or metastatic c-KIT expressing adenoid cystic carcinoma and for nonadenoid cystic malignant salivary tumors

BACKGROUND

Adenoid cystic carcinoma (ACC) is a subtype of malignant salivary gland tumors (MSGT), in which 90% of cases express cKIT. Dasatinib is a potent and selective inhibitor of five oncogenic protein tyrosine kinases (PTKs)/kinase families including cKIT. We conducted a phase II study to determine the antitumor activity of dasatinib in ACC and non-ACC MSGT.

PATIENTS AND METHODS

In a two-stage design, patients with progressive, recurrent/metastatic ACC (+cKIT) and non-ACC MSGT (separate cohort) were treated with dasatinib 70 mg p.o. b.i.d. Response was assessed every 8 weeks using RECIST.

RESULTS

Of 54 patients: 40 ACC, 14 non-ACC (1, ineligible excluded); M:F = 28 : 26, median age 56 years (range 20-82 years), ECOG performance status 0 : 1 : 2 = 24 : 28 : 2, prior radiation: 44, prior chemotherapy: 21. The most frequent adverse events (AEs) (as % of patients, worst grade 2 or higher) were: fatigue (28%), nausea (19%), headache (15%), lymphopenia (7%), dyspnea (11%), alanine aminotransferase increased (7%), anorexia (7%), vomiting (7%), alkaline phosphatase increased (6%), diarrhea (6%), neutropenia (6%), and noncardiac chest pain (6%). No grade 4 AE occurred, 15 patients experienced a grade 3 AE, primarily dyspnea (5) and fatigue (4), and cardiac toxicity (1 prolonged QTc). Among ACC patients, best response to dasatinib: 1 patient (2.5%) had partial response, 20 patients (50%) had stable disease (SD) (3-14 months), 12 patients (30%) had PD, 2 withdrew, 3 discontinued therapy due to AE, and 2 died before cycle 2. Median progression-free survival was 4.8 months. Median overall survival was 14.5 months. For 14 assessable non-ACC patients, none had objective response, triggering early stopping rule. Seven had SD (range 1-7 months), 4 PD, 2 discontinued therapy due to AE, and 1 died before cycle 2.

CONCLUSION

Although there was only one objective response, dasatinib is well tolerated, with tumor stabilization achieved by 50% of ACC patients. Dasatinib demonstrated no activity in non-ACC MSGT.

Genetic variants of kinase suppressors of Ras (KSR1) to predict survival in patients with ERα-positive advanced breast cancer

In patients with breast cancer (BC), deregulation of estrogen receptor (ERα) activity may account for most resistance to endocrine therapies. Our previous study used a whole-human kinome siRNA screen to identify functional actors in ERα modulation and showed the implication of proteins kinase suppressors of ras (KSR1). From those findings we evaluated the clinical impact of KSR1 variants in patients with ERα+ BC treated with TAM. DNA was obtained from 222 patients with advanced ERα+ BC treated with TAM who had undergone surgery from 1981 to 2003. We selected three potentially functional relevant KSR1 polymorphisms; two within the 3'UTR (rs224190, rs1075952) and one in the coding exon 7 (rs2293180). The primary end points were overall survival (OS) and disease-free survival (DFS). After a 6.4-year median follow-up, patients carrying the rs2241906 TT genotype showed shorter DFS (2.1 vs 7.1 years, P=0.005) and OS (2.6 vs 8.4 years P=0.002) than those with the TC or TT genotypes. Those associations remained significant in the multivariable analysis adjusting age, lymph node status, LMTK3 and IGFR variants and HER2 status. The polymorphisms rs2241906 and rs1075952 were in linkage disequilibrium. No association was shown between rs2293180 and survival. Among the actors of ERα signaling, KSR1 rs2241906 variants may predict survival in patients with advanced ERα+ BC treated with adjuvant TAM.

KSR1 regulates BRCA1 degradation and inhibits breast cancer growth

Kinase suppressor of Ras-1 (KSR1) facilitates signal transduction in Ras-dependent cancers, including pancreatic and lung carcinomas but its role in breast cancer has not been well studied. Here, we demonstrate for the first time it functions as a tumor suppressor in breast cancer in contrast to data in other tumors. Breast cancer patients (n>1000) with high KSR1 showed better disease-free and overall survival, results also supported by Oncomine analyses, microarray data (n=2878) and genomic data from paired tumor and cell-free DNA samples revealing loss of heterozygosity. KSR1 expression is associated with high breast cancer 1, early onset (BRCA1), high BRCA1-associated ring domain 1 (BARD1) and checkpoint kinase 1 (Chk1) levels. Phospho-profiling of major components of the canonical Ras-RAF-mitogen-activated protein kinases pathway showed no significant changes after KSR1 overexpression or silencing. Moreover, KSR1 stably transfected cells formed fewer and smaller size colonies compared to the parental ones, while in vivo mouse model also demonstrated that the growth of xenograft tumors overexpressing KSR1 was inhibited. The tumor suppressive action of KSR1 is BRCA1 dependent shown by 3D-matrigel and soft agar assays. KSR1 stabilizes BRCA1 protein levels by reducing BRCA1 ubiquitination through increasing BARD1 abundance. These data link these proteins in a continuum with clinical relevance and position KSR1 in the major oncoprotein pathways in breast tumorigenesis.

KRAS mutations in non-small-cell lung cancer and colorectal cancer: implications for EGFR-targeted therapies

BACKGROUND

KRAS mutations are associated with diverse biologic functions as well as prognostic and predictive impact in non-small cell-lung cancer (NSCLC) and colorectal cancer (CRC). In CRC, benefit from monoclonal antibody therapies targeting EGFR is generally limited to patients whose tumors have wild-type (WT) KRAS, whereas data suggest that this association is not present for NSCLC. We hypothesized that the unique tobacco-related carcinogenesis of NSCLC results in a divergence of KRAS MT genotype compared with CRC, contributing to differences in outcomes from EGFR-targeted therapies.

MATERIAL AND METHODS

Tumor from 2603 patients (838 CRC and 1765 NSCLC) was analyzed for KRAS mutations. DNA was extracted from microdissected formalin-fixed-paraffin-embedded specimens (FFPE) and 7 different base substitutions in codons 12 and 13 of KRAS were determined.

RESULTS

KRAS mutation genotype differed significantly between NSCLC and CRC in frequency (25% vs. 39%; p<0.001), smoking-associated G>T transversions (73% versus 27%; p<0.001), and ratio of transversions to transitions (3.5 vs. 0.79; p<0.001). In NSCLC GLY12Cys mutations, resulting from a codon 12 GGT>TGT substitution, were observed in 44% compared to 10% for CRC. In contrast, codon 12 or 13 GLY>ASP substitutions (resulting in a G>A transition) were more frequent in CRC (42%) compared with NSCLC (21%).

CONCLUSION

In this large dataset, KRAS mutation patterns are quantitatively and qualitatively distinct between NSCLC and CRC, reflecting in part differences in tobacco-related carcinogenesis. In light of differences in predictive value for EGFR-directed monoclonal antibody therapy and prognosis for specific KRAS mutations between NSCLC and CRC, these data provide an underlying biologic rationale.

Assessment of HER2 gene amplification in adenocarcinomas of the stomach or gastroesophageal junction in the INT-0116/SWOG9008 clinical trial

BACKGROUND

Trastuzumab has been approved for patients with human epidermal growth factor receptor 2 (HER2)-positive metastatic gastric carcinoma; however, relatively little is known about the role of HER2 in the natural history of this disease.

PATIENTS AND METHODS

Patients enrolled in the INT-0116/SWOG9008 phase III gastric cancer clinical trial with available tissue specimens were retrospectively evaluated for HER2 gene amplification by FISH and overexpression by immunohistochemistry (IHC). The original trial was designed to evaluate the benefit of postoperative chemoradiation compared with surgery alone.

RESULTS

HER2 gene amplification rate by FISH was 10.9% among 258 patients evaluated. HER2 overexpression rate by IHC was 12.2% among 148 patients evaluated, with 90% agreement between FISH and IHC. There was a significant interaction between HER2 amplification and treatment with respect to both disease-free survival (DFS) (P = 0.020) and overall survival (OS) (P = 0.034). Among patients with HER2-non-amplified cancers, treated patients had a median OS of 44 months compared with 24 months in the surgery-only arm (P = 0.003). Among patients with HER2-amplified cancers, there was no significant difference in survival based on treatment arm. HER2 status was not a prognostic marker among patients who received no postoperative chemoradiation.

CONCLUSION

Patients lacking HER2 amplification benefited from treatment as indicated by both DFS and OS.

CLINICAL TRIAL

INT-0116/SWOG9008 phase III.

Assessment of HER2 gene amplification in adenocarcinomas of the stomach or gastroesophageal junction in the INT-0116/SWOG9008 clinical trial

BACKGROUND

Trastuzumab has been approved for patients with human epidermal growth factor receptor 2 (HER2)-positive metastatic gastric carcinoma; however, relatively little is known about the role of HER2 in the natural history of this disease.

PATIENTS AND METHODS

Patients enrolled in the INT-0116/SWOG9008 phase III gastric cancer clinical trial with available tissue specimens were retrospectively evaluated for HER2 gene amplification by FISH and overexpression by immunohistochemistry (IHC). The original trial was designed to evaluate the benefit of postoperative chemoradiation compared with surgery alone.

RESULTS

HER2 gene amplification rate by FISH was 10.9% among 258 patients evaluated. HER2 overexpression rate by IHC was 12.2% among 148 patients evaluated, with 90% agreement between FISH and IHC. There was a significant interaction between HER2 amplification and treatment with respect to both disease-free survival (DFS) (P = 0.020) and overall survival (OS) (P = 0.034). Among patients with HER2-non-amplified cancers, treated patients had a median OS of 44 months compared with 24 months in the surgery-only arm (P = 0.003). Among patients with HER2-amplified cancers, there was no significant difference in survival based on treatment arm. HER2 status was not a prognostic marker among patients who received no postoperative chemoradiation.

CONCLUSION

Patients lacking HER2 amplification benefited from treatment as indicated by both DFS and OS.

CLINICAL TRIAL

INT-0116/SWOG9008 phase III.

Selecting the best targeted agent in first-line treatment of unresectable liver metastases from colorectal cancer: does the bench have the answers?

For physicians facing patients with organ-limited metastases from colorectal cancer, tumor shrinkage and sterilization of micrometastatic disease is the main goal, giving the opportunity for secondary surgical resection. At the same time, for the majority of patients who will not achieve a sufficient tumor response, disease control remains the predominant objective. Since FOLFOX or FOLFIRI have similar efficacies, the challenge is to define which could be the most effective targeted agent (anti-EGFR or anti-VEGF) to reach these goals. Therefore, a priori molecular identification of patients that could benefit from anti-EGFR or anti-VEGF monoclonal antibodies (i.e. the currently approved targeted therapies for metastatic colorectal cancer) is of critical importance. In this setting, the KRAS mutation status was the first identified predictive marker of response to anti-EGFR therapy. Since it has been demonstrated that tumors with KRAS mutation do not respond to anti-EGFR therapy, KRAS status must be determined prior to treatment. Thus, for KRAS wild-type patients, the choices that remain are either anti-VEGF or anti-EGFR. In this review, we present the most updated data from translational research programs dealing with the identification of biomarkers for response to targeted therapies.

Southwest Oncology Group study S0413: a phase II trial of lapatinib (GW572016) as first-line therapy in patients with advanced or metastatic gastric cancer

BACKGROUND

Lapatinib (GW572016) is a dual tyrosine kinase inhibitor of epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2/ErbB2), which are reported as overexpressed in 15%-45% of gastric cancers, making them potential targets.

PATIENTS AND METHODS

The primary objective of this study was to assess response rate. Secondary objectives included overall survival (OS), toxicity, and the relationship of EGFR, ErbB2, and markers of angiogenesis with clinical outcome. Lapatinib was administered to chemonaive metastatic gastric cancer patients at a dose of 1500 mg orally daily for 28 days.

RESULTS

The study enrolled 47 patients from February 2005 until May 2006. Four patients (9%) had a confirmed partial response (PR), 1 (2%) had an unconfirmed PR, and 10 (23%) had stable disease. Median (95% confidence interval) time to treatment failure was 1.9 (1.6-3.1) months and OS was 4.8 (3.2-7.4) months. Significant adverse events: one grade 4 cardiac ischemia/infarction, one grade 4 fatigue, and one grade 4 emesis. One treatment-related death was due to central nervous system ischemia. An exploratory analysis of markers revealed gene expression of HER2, interleukin (IL)-8 and genomic polymorphisms IL-8, and vascular endothelial growth factor correlated with OS.

CONCLUSIONS

Lapatinib is well tolerated, with modest single-agent activity in advanced/metastatic gastric cancer patients. Potential molecular correlatives were identified which warrant further validation.

Association of gender, age, and ethnicity with survival in patients with pancreas cancer

e15587

Background: Pancreatic cancer (PC) is the fourth leading cause of cancer death in the United States. However little is still known about factors that influence its development and progression. Recent data suggest that PC is, at least in part, an estrogen- dependent disease; there is growing epidemiological evidence that aspects of reproductive history and hormonal exposure are associated with risk of this disease. It was shown that age at menarche of <13 is associated with less risk of PC. However no data are available whether gender is associated with outcome in patients with PC. The purpose of this study was to test whether age, gender or ethnicity influence the outcome in PC. Methods: The data of the 50,302 adults diagnosed with PC between 1988 and 2004 were extracted from the Surveillance Epidemiology and End Results public use database. These included 24,240 patients diagnosed with localized pancreatic cancer (LPC) and 26,062 patients with metastatic pancreatic cancer (MPC). Demographic, clinical variables and survival time were retrieved. The primary endpoint was overall survival. We constructed Cox proportional hazards models to evaluate association between patient characteristics and survival in LPC and MPC separately. Pair interactions were also tested. Results: On multivariate analysis gender, age, race, marital status, tumor size, grade, histology, type of treatment and lymph node involvement were found to be independent predictors of survival. Females had a significant longer survival, with an HR of 0.959 (95% CI: 0.932–0.987) among patients with LPC and an HR of 0.918 (95%CI: 0.894–0.942) among patients with MPC. Each age group displayed a significant longer survival than its correspondent older age group. When we combined age and gender in the analysis, females had a longer survival than males in each single age group in the MPC group. In the LPC group the longer survival of female patients was only observed in the youngest age group. Conclusions: This is the first and largest study to address gender and outcome in PC. Our data suggest that the estrogen pathway may play an important prognostic role in patient with this disease. These data also warrant further in vitro and in vivo investigations on the mechanisms of estrogen and pancreas progression.

No significant financial relationships to disclose.

Wt Kras and gene expression levels of VEGFR2, EGFR, and ERCC-1 associated with progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC) treated with first-line 5-FU or capecitabine with oxaliplatin and bevacizumab (FOLFOX/BV or XELOX/BV)

4056

Background: While wild type (wt) Kras is associated with improved outcome to anti-EGFR therapy in pts with mCRC, there are no identified predictors of outcome for FOLFOX/BV. We evaluated Kras status and expression of genes involved in angiogenesis, DNA repair and 5-FU metabolism in 68 patients treated with FOLFOX/BV or XELOX/BV. These genes included VEGF, VEGF-receptor 2 (KDR), Cox-2, IL 6 and 8, chemokine-receptors 1 & 2, EGFR and ERCC-1. Methods: Tissue samples from 68 patients with mCRC were analyzed. mRNA was extracted from laser-capture-microdissected tumor tissue. cDNA was prepared by reverse transcription and quantitation of the candidate genes was performed using a fluorescence- based real-time detection method (TaqMan). Allele specific RT-PCR was performed to determine Kras mutation status in codons 12 and 13. Results: There were 68 pts (38 males, 30 females), median age: 56 years (range 29–81). All received first line 5FU, oxaliplatin and BV (28 FOLFOX/BV, 40 XELOX/BV). Radiologic response: 1 CR, 39/68 (57%) PR, 27/68 (40%) SD, and 1 PD. Median OS is not reached. At a median follow-up of 32.0 months (mo) (range: 2.3–47.8 mo), the median PFS was 12.4 mo (95% CI: 9.8–15.2). Kras mutation was identified in 39 pts (57%). RR was 64% in pts with wt Kras and 52% in pts with mutant Kras (p=0.33). PFS was significantly longer for pts with wt kras compared to pts with mutant kras (13.7 mo [95% CI: 6.9–13.2] versus 8.3 mo [95%CI: 6.9–13.2], P=0.039). High EGFR (median PFS: 15.2 mo; 95% CI 11.7–16.5 mo), high VEGFR2 (median PFS: 13.9 mo; 95% CI 11.0–16.5 mo), and low ERCC1 (median PFS: 12.4 mo; 95% CI 10.9–16.4 mo) were associated with longer PFS compared to low EGFR (median PFS: 7.9 mo; 95% CI 6.9–11.0 mo, P=0.040), low VEGFR2 (median PFS: 7.2 mo; 95% CI 6.5–8.1 mo, P=0.032), and high ERCC1 (median PFS: 9.6 mo; 95% CI 5.8–15.2 mo, P=0.045). Conclusions: To our knowledge, this is the first report of a potential association between Kras status as well as gene expression levels of VEGFR2, ERCC-1 and EGFR and clinical outcome to FOLFOX/BV therapy in pts with mCRC. Prospective clinical trials are needed to validate these results.

[Table: see text]

Cisplatin + irinotecan versus cisplatin + etoposide in extensive stage small cell lung cancer (E-SCLC): Final “common arm”: Comparative outcomes analysis of JCOG 9511 and SWOG 0124

8027

Background: S0124 was a large North American phase III trial (n=651) that failed to confirm a survival benefit for cisplatin/irinotecan over cisplatin/etoposide in patients with E-SCLC, contrary to the results of J9511, a phase III trial exclusively in Japanese patients (n=154). As S0124 and J9511 protocols used identical treatment regimens and similar eligibility criteria, we compared demographics, toxicity, and outcomes using patient-level data and a “common arm” analysis to explore potential reasons for the divergent results. Methods: In both trials, patients with documented E-SCLC and adequate end-organ function were randomized to receive either cisplatin 60 mg/m2 day 1 + irinotecan 60 mg/m2 days 1, 8, & 15 Q 4 weeks or cisplatin 80 mg/m2 day 1 + etoposide 100 mg/m2 days 1–3 Q 3 weeks. Demographics and outcomes data were compared among 805 patients enrolled in J9511 and S0124 receiving identical treatment using a logistic model adjusted for age, sex, and performance status. Results: Of 671 patients in S0124, 651 were eligible. Patient characteristics (J9511 & S0124, respectively): Mean age - 61 & 62 years; Male sex - 132 (86%) & 370 (57%), p<0.001; Performance status 0 - 19 (12%) & 211 (32%), p<0.001. Efficacy and toxicity comparisons are summarized below. Conclusions: Significant differences in patient demographics, toxicity, and efficacy exist between J9511 and S0124 populations. These results, relevant in the current era of clinical trials globalization, warrant 1) consideration of differential patient characteristics and outcomes amongst populations receiving identical therapy; 2) utilization of the “common arm” model in prospective trials; and 3) inclusion of pharmacogenomic correlates in cancer trials where ethnic/racial differences in drug disposition are expected.

[Table: see text]

[Table: see text]

Effect of gender and age on overall survival in patients with esophageal cancer

4541

Background: The American Cancer Society estimated that during 2008 approximately 16,470 new esophageal cancer cases would be diagnosed in the United States. Previous studies indicate that the incidence of esophageal cancer is more in males than females; however the influence of sex in the progression of esophageal cancer is not clearly understood. In vitro and in vivo models showed that administration of estradiol significantly inhibited the growth of ER-positive and AR-positive KSE-1 tumors in both males and females in conjunction with an increase in the estradiol levels and a decrease in the DHT levels in the serum. Thus we tested the clinical significance of sex in the overall survival of esophageal cancer using SEER data. Methods: A total of 21,584 patients with localized and metastatic esophageal cancer from 1988–2004 were screened using the SEER registry. The patients were divided into age at diagnosis, sex and ethnicity. The analysis for overall survival was based on the Cox proportional hazards model adjusted for marital status, site of primary tumor, treatment, histology and tumor grade and stratified by year of diagnosis and SEER registry site. Pairwise interactions (age and sex, age and race, and sex and race) were also examined. Results: Females with localized esophageal cancer had significantly longer overall survival compared to males (p<0.001). In metastatic esophageal cancer, females seemed to have longer overall survival than males (p=0.054). Overall survival decreased with increasing age (p<0.001). African Americans with localized esophageal cancer have worse median overall survival compared to Caucasians (p<0.001). No significant difference was noted with respect to ethnicity in metastatic esophageal cancer (p=0.22). Also pairwise interactions did not have significant difference. Conclusions: This is the first and largest study showing gender as an independent prognostic factor in patients with localized and metastatic esophageal cancer. Females had a significant better overall survival than males in esophageal cancer suggesting that sex hormone pathways may have a potential impact on tumor progression. These data warrant further studies to explore the role of these pathways in the diagnosis and treatment of esophageal cancer.

No significant financial relationships to disclose.

Association of CD133 polymorphisms and clinical outcome in metastatic colorectal cancer (mCRC) patients (pts) treated with either first-line 5-FU + bevacizumab (BV) or second-line irinotecan (IR)/cetuximab (CB) or IR alone

4062

Background: CD133 has been routinely used to identify colon cancer stem cells. A recent study indicated that elevated levels of CD133 plasma mRNA correlated with colon cancer recurrence. Furthermore plasma levels of CD133+ progenitor cells have been found to be decreased after treatment with BV. We tested whether potentially functional frequently occurring germline variations in the 3’UTR-region of the CD133 gene (rs2240688, rs3130 and rs2286455), might be associated with clinical outcome in first- and second-line treated mCRC pts. Methods: Genomic DNA was extracted either from peripheral blood (79 pts, who were enrolled in a phase-II clinical trial with FOLFOX/BV or XELOX/BV) or formalin-fixed paraffin-embedded tumor samples (186 pts, who were enrolled in the EPIC phase III clinical trial, US-sites only) of mCRC pts. Pts received either first-line treatment with FOLFOX/ BV (33 pts) or XELOX/BV (46 pts) or second-line treatment with CB/IR (84 pts, arm A) or IR (102 pts, arm B) alone. Genotyping was performed using PCR-RFLP assays. Results: 79 pts (47 men, 32 women) received FOLFOX/BV or XELOX/BV. Radiologic response: 43 pts (54%) CR/PR, 35 pts (45%) SD/PD. Median PFS was 10.8 months (95%CI: 8.1–14.9). The second cohort consisted of 186 pts (103 men, 83 women). Radiologic response: Arm A 11 pts (13%) CR/PR, 73 pts (87%) SD/PD. Arm B 6 pts (6%) CR/PR, 96 pts (94%) SD/PD. Median PFS (arm A) was 3.0 months (95%CI: 2.4–4.1) vs. 2.7 months (arm B,95%CI: 2.2–2.9). Combined analysis of rs2286455 and rs3130 showed a significant association with PFS (p= 0.010, log-rank test) in pts receiving FOLFOX/BV or XELOX/BV. In pts receiving IR alone rs2240688 was significantly associated with OS (p=0.0128, log-rank test). Multivariate analysis showed a significant association with PFS in first-line setting for rs2286455 and rs3130 (adjusted p=0.012) and a trend in second-line setting for rs2240688 (adjusted p=0.086). Conclusions: These are the first data to show that polymorphisms in CD133 predict outcome in mCRC pts in first- and second- line setting, suggesting that CD133 may be a potential predictive marker. These results need to be confirmed in larger prospective studies.

[Table: see text]

Molecular profiling in patients (pts) with upper gastrointestinal (UGI) cancers correlated with clinical outcome

e22042

Background: Recent efforts have expanded our understanding of the molecular pathogenesis of UGI cancers and how oncogenes and tumor suppressor genes play a role in both carcinogenic and metastatic processes. These markers may serve as prognostic and/or predictive factors for recurrence following resection and resistance to radiotherapy and chemotherapy. Gene expression levels of thymidylate synthase (TS), thymidylate phosphorylase (TP), epidermal growth factor receptor (EGFR) and excision repair cross complementing (ERCC-1) have been shown to be associated with outcomes in lung, gastric and colon cancer. We evaluated TS, ERCC-1, EGFR and TP gene expression levels in patients with UGI cancers. Methods: This was a retrospective study of 80 pts with UGI cancers evaluated at USC who underwent molecular profiling. The primary objective was to determine a correlation between TS, TP, ERCC1 and EGFR and correlate with clinical outcome. Characteristics of these 80 pts: (16 females, 64 males ); median age 61 years (range 34–85); tumor types evaluated - 32 (40%) esophageal, 24(30%) gastric, 24(30%) GE junction; stages- 53 % IV, 21% III, 12% II, 2% I were evaluated for intratumoral gene expression of TS, TP, ERCC-1, & EGFR by real time quantitative PCR using Taqman technology from microdisected paraffin-embedded tumor sections. Results: High TS expression was associated with shorter OS (12.8 months vs. 23.7 months p=0.036). A significant correlation was found between TP & ERCC-1 (p=0.0078, r=0.37); TP & TS (p=0.0128, r=0.35); TP & EGFR (p=0.0065, r=0.39); TS & ERCC-1 (p=0.0004, r=0.39); ERCC-1 & EGFR (p=0.025, r=0.30). No statistically significant relationship was found between TS & EGFR (p=0.06,r=0.25). There was no correlation of ERCC-1, TP, & EGFR with OS that reached statistical significance. Conclusions: TS mRNA levels were shown to be associated with OS in UGI tumors, consistent with data reported in colon cancer. TS gene expression was significantly associated with expression levels of ERCC-1. In addition, ERCC1 was associated with EGFR. These data show for the first time that molecular pathways of cytotoxic agents are linked to the EGFR pathway suggesting that sensitivity to fluoropyrimidines, oxaliplatin, and EGFR inhibitors may be associated.

No significant financial relationships to disclose.

Phase II study of capecitabine and gemcitabine in patients with metastatic colorectal cancer (mCRC)

e15077

Background: Failing standard chemotherapy, many mCRC patients (pts) maintain an adequate performance status necessitating therapeutic options. Preclinical data in CRC cell lines shows that 5-FU and gemcitabine have synergistic cytotoxicity by stabilization of thymidylate synthase resulting in stronger inhibition of DNA synthesis. Methods: The primary objective was to assess time to progression (TTP) in mCRC pts who had progressed on irinotecan and oxaliplatin therapy. Secondary objectives included response rate (RR), overall survival (OS), toxicity and correlates to assess genes in the 5-FU, gemcitabine pathways. This single center study was a two stage minimax design. A ≤ 30% chance of progressing within the first 6 weeks would be promising. 27 pts were enrolled in the first stage, if ≥ 13 progressed (or had unacceptable toxicity) prior to 6 weeks then the study would be terminated. The planned sample size was 53 pts. A cycle was defined as capecitabine 650 mg/m2 BID days 1–14 and gemcitabine 1,000 mg/m2 i.v. over 100 minutes days 1, 8. Results: The study met its first stage goal and continued to accrue 54 eligible pts (male 24/female 30), median age 58 years (range 31–78). Pts received a median of 2 cycles (range 1–19), with 22% of pts receiving 6 + cycles. Three pts that withdrew shortly after start were excluded from toxicity analysis. Of 51 pts, 23 had Grade 3/4 drug related toxicity, commonly 9 pts ANC/AGC; 4 pts pain. The probability of progressing at 6 weeks was .46 + .07, with 23 pts progressing or dying within that time. Median OS was 5.9 months (95% CI 4.3, 8.7), PFS was 8.1 weeks (95% CI 5.6, 11.6). Of 48 pts evaluable for response, 17 had stable disease, for these pts, the median PFS was 4.3 months (95%CI: 3.9–8.8) and OS 10 months (95%CI: 7.9–16.6); 31 had progressive disease or symptomatic deterioration. Conclusions: The combination of gemcitabine and capecitabine in refractory mCRC was well tolerated. Unfortunately, the study did not meet the primary endpoint, but there appears to be a subset of pts with stable disease who may benefit from therapy. Correlatives are pending.

[Table: see text]

Polymorphisms in epidermal growth factor (EGF) and proteinase activated receptor 1 (PAR-1) associated with tumor recurrence in localized adenocarcinoma (EA) of the esophagus treated with surgery alone

4564

Background: Tumor angiogenesis is a well-recognized aspect of human cancer biology and is mediated at least in part by EGF and PAR-1, which in turn may impact the process of tumor growth and progression. Systemic tumor recurrence after curative resection continues to be a significant problem in the management of patients with localized EA. Further, it is being increasingly recognized that esophageal squamous cell carcinoma and EA are separate and distinct disease groups and need to be considered individually. We therefore designed a large retrospective study of EA patients to identify novel molecular markers of prognosis to better define tumor stage and progression, and help to define novel targets, as well as surrogate-endpoints of disease progression and response to therapy. Methods: Between 1992 and 2005 normal esophageal tissue samples from 239 patients with localized EA treated with surgery alone were obtained at University of Southern California medical facilities. The median follow-up was 3.2 years. 114 out of 239 (48%) patients had tumor recurrence, with a probability of 5-year recurrence of 0.62 ± 0.04. DNA was isolated from formalin-fixed paraffin-embedded specimens and 10 angiogenesis related and functional gene polymorphisms were analyzed using a PCR-RFLP and 5´-end [γ-33P] ATP-labeled PCR method. Results: PAR-1 -506 ins/del (p-value=0.003; log-rank test) and EGF +61 A>G (p-value=0.034; log-rank test) are adverse prognostic markers in univariate analysis. After adjusting for covariates (gender, T1-, N-category, type of surgery) in the multivariable model, "high-expression" variants of PAR-1 (any insertion allele) (RR: 1.81; adjusted p-value = 0.011) and EGF (A/A) (RR: 1; adjusted p-value=0.035) remained significantly associated with time to recurrence, compared to other genotype combinations of PAR-1 (RR: 1) and EGF (RR: 0.65). Conclusions: This study supports the role of functional EGF and PAR-1 polymorphisms as independent prognostic markers in localized EA and may therefore help to identify patient subgroups at high risk for tumor recurrence. Prospective and biomarker-embedded clinical trials are needed to validate our findings.

[Table: see text]

SWOG 0514: A phase II study of sorafenib (BAY 43–9006) as single agent in patients (pts) with unresectable or metastatic gallbladder cancer or cholangiocarcinomas

4639

Background: There are no standard chemotherapeutic regimens for incurable biliary adenocarcinomas. Gemcitabine and fluoropyrimidine- based chemotherapy results in occasional responses and a median survival approaching 6 months. Given the presence of b- raf mutations and overexpression of VEGF in biliary cancers, we initiated a study to evaluate the efficacy of sorafenib, a multitargeted tyrosine kinase inhibitor of c and b-Raf, VEGFR-2/3 and PDGFR, in pts with metastatic biliary cancers. Methods: Pt eligibility included adequate organ function and no prior treatment for metastatic disease. The primary end-point was objective response rate (RR). Secondary endpoints were overall survival (OS) and progression free survival (PFS). A two-stage design was used to detect a difference in the null hypothesis of 5% response probability and the alternative 20% response probability. If at least one confirmed response occurred after the first 25 pts, another 25 were to be accrued. Sorafenib was administered at 400 mg PO BID continuously. A cycle was defined as 28 days of therapy, and radiological assessment was done every 2 cycles. Results: 36 pts were enrolled during the first stage of accrual. 5 pts were ineligible. 52% were female. Median age was 57.8 years (range 33.8–81.5). Adverse events: 1 pt died with grade 4 supraventricular tachycardia and venous thromboembolism. Grade 3/4 toxicities were noted in 20 pts (66.7%) and included hand-foot syndrome in 4 pts (13%), while thrombosis/embolism, elevated liver transaminases, and abdominal pain were each seen in 3 pts (10% for each). Reversible posterior leukoencephalopathy syndrome, GI perforation, and GI hemorrhage were each seen in 1 pt (3% for each). 2 pts (6%) had an unconfirmed partial response and 9 pts (29%) had stable disease. 27 pts have progressed. Median PFS was 2 months (95% CI: 2–4 months). 14 pts have died, with a median survival estimate of 6 months (95% CI: 4–10 months). Conclusions: Sorafenib did not result in a clinically significant objective RR in pts with gallbladder and cholangiocarcinoma but demonstrated an impact on survival that may be comparable to commonly used chemotherapy regimens.

No significant financial relationships to disclose.

How well do we communicate with patients concerning adjuvant systemic therapy? A survey of 150 colorectal cancer survivors

4020

Background: Adjuvant chemotherapy (AC) presents a substantial patient education challenge to medical oncologists (MOs). Findings from our 2005 pilot survey suggest that a significant fraction of colorectal cancer (CRC) survivors are willing to undergo AC for modest treatment benefits, but their understanding of risks and benefits may be suboptimal. This project attempted to validate these findings by surveying patients with CRC (Pts) who previously received AC. Methods: 150 Pts who received AC for CRC in the last 5 years were recruited to listen to an audio program on AC featuring interviews with clinical investigators (CIs) and Pts who received AC. Based on this input, Pts were asked whether they would undergo the same AC again for varying absolute treatment benefits. A corresponding survey asked 24 CRC CIs and 150 MOs to predict how patients would respond. The survey also queried Pts about their expectations of and experiences with AC side effects. Results: About 1/3 of Pts would be treated again with AC for a 1% absolute reduction in recurrence risk (ARRR), and about 2/3 believed a 5% ARRR would justify treatment. There were no statistically significant differences between responses of males and females or between Pts receiving oxaliplatin (OX) and those receiving other regimens. ( Table 1 ) The corresponding estimates of CIs and MOs were lower. Additionally, AC side effects were different than expected: 57% and 66% of Pts experienced less GI toxicity and alopecia, respectively, while 38% and 46% of Pts receiving OX experienced greater cold intolerance and numbness. Conclusions: Many potential obstacles exist in communicating with Pts about AC including heterogeneity in Pts’ attitudes towards risk/benefit trade-offs and preconceptions about treatment side effects. The next step in this initiative will be to examine these issues prospectively by evaluating the impact of an audio/web education supplement on the decision-making process.

[Table: see text]

No significant financial relationships to disclose.

High plasma levels of fenretinide (4-HPR) were associated with improved outcome in a phase II study of recurrent ovarian cancer: A study by the California Cancer Consortium

5555

Background: 4-HPR is a synthetic retinoid with dose-dependent cytotoxicity in vitro for ovarian cancer cell lines, especially at concentrations > 10 μM. We previously reported preliminary results of a phase II trial (Garcia et al, Proc ASCO 2004: a5056) that showed encouraging stabilization of disease. Methods: 4-HPR capsules were given orally (900 mg/m2/bid for 7 days every 21-days) to women with recurrent ovarian cancer. Steady-state plasma concentrations of 4-HPR were determined by HPLC. Patients were required to have uni-dimensionally measurable disease, 0–2 chemotherapy regimens for recurrent disease, and performance status 0–2. The primary endpoint was response rate by RECIST criteria; secondary endpoints were progression-free survival (PFS) and overall survival (OS). Results: 31 patients were enrolled to the first stage; 28 patients were evaluable for response, with a median of 3 prior chemotherapy regimens (range: 1–3); 24 patients were evaluable for pharmacokinetics. A total of 172 4-HPR treatment cycles were given. Eleven patients experienced at least one grade 3 toxicity and one patient had grade 4 fatigue. No objective responses were observed, leading to study closure. Best response was stable disease in 13 patients (42%) with a median duration of 7.9 months (2.1–25.8 months). 4-HPR steady-state plasma levels ranged from 3.1–12.5 μM. PFS at 6 months was 26% (± 0.08 SE), 42 % (± 0.14 SE) for patients with 4-HPR plasma levels = 9 μM (n=12), and 17% (± 0.11 SE) for plasma levels < 9 μM (n=12). OS with plasma levels = 9 μM was significantly higher than for those with 4-HPR levels < 9 μM (66% vs. 13% at 18 months, adjusted p-value = 0.016). Conclusions: 4-HPR may provide clinical benefit to patients with recurrent ovarian cancer who obtain high 4-HPR plasma levels. Trials in ovarian cancer of 4-HPR focused on progression-free survival as an endpoint are justified using formulations with improved bioavailability. Supported by N01-CM-62209.

No significant financial relationships to disclose.

Heparanase and Survivin—potential markers of aggressiveness of colon cancers

14524

Background: Survivin is an inhibitor of apoptosis and specifically expressed in several human cancers. Heparanase seems to play an important role in cancer metastasis, angiogenesis and regulation of COX-2 expression. The purpose of this pilot study was to evaluate whether there is a difference in the level of survivin and heparanase expression in paraffin-embedded tissue from patients with colon cancer with and without metastasis. Methods: Paraffin-embedded tissue samples were obtained from 32 patients with primary colon tumors with (n = 15) and without liver metastases (n = 17) at the time of diagnosis. After laser capture microdissection direct quantitative real-time reverse transcriptase PCR (RT-PCR, TaqMan™) assays were performed in triplicates to determine survivin and heparanase gene expression levels. Gene expression was normalized with beta-Actin. Decision tree analysis was used to assess whether higher levels of these genes are associated with the existence of metastasis. Results: Neither survivin nor heparanase gene expression alone showed a significant relationship with the pM-Stage, although the detection levels of heparanase showed a tendency towards significant correlation with the existence of liver metastasis (p = 0.06). Decision tree analysis was used to split patients into two groups based on cut-off values of gene expression most closely associated with the pM-stage. The groupings determined by this analysis were significantly differently correlated with metastasis by Spearman’s test (p=0.03). Conclusions: Our results show that high expression levels of both survivin and heparanase in paraffin- embedded tissue are significantly correlated with the pM-Stage in this patient cohort. In conclusion, using intratumoral survivin and heparanase gene expression levels might be useful to identify patients at risk for liver metastasis. Prospective and larger clinical studies are warranted to validate our preliminary findings.

No significant financial relationships to disclose.

Use of sex-specific genomic profile to predict clinical outcome in metastatic colorectal cancer (mCRC) patients treated with 5- fluorouracil/oxaliplatin

4122

Background: There is increasing evidence that gender plays a significant role in the development and progression of colorectal cancer (CRC). Rates of CRC incidence are higher among males, and it has been shown that hormone replacement therapy in postmenopausal women reduces the risk of developing CRC, indicating a protective effect of female hormones. These findings indicate that physiological differences between the sexes may contribute to differential tumor development and progression. In the current study, we tested the hypothesis whether males and females would have different genomic profiles that would predict clinical outcome in 5-FU/oxal-treated mCRC. Methods: One hundred seventy-three patients were enrolled in this phase II study. One hundred fifty-two patients were evaluable for genotyping and statistical analysis. There were 74 females and 78 males, and median age was 60 (range 25–87). Twenty-four polymorphisms in 12 genes involved in cancer progression were tested by PCR. CART analysis was used to test for association of polymorphisms with overall survival (OS), time to tumor progression (TTP), and tumor response. Separate regression trees were developed for males and females. Results: Median follow-up was 18.6 months, response rate 19%, median time to tumor progression 4.2 months and median survival 10.3 months. CART analysis by gender generated significantly different regression trees for men versus women. For TTP, males carrying polymorphisms in the ER-β and SCN1A genes had improved TTP, while females carrying polymorphisms in the XPD and EGFR genes had improved TTP. For OS, male patients carrying ER-β and MTHFR polymorphisms had improved OS, while female patients carrying polymorphisms in SCN1A and PLA2 had improved OS. It is notable that analysis of males and females in combination produced a different regression tree than when separated by gender. Conclusions: Our results suggest for the first time that genomic profiling to predict clinical outcome of patients with mCRC depends on gender. Taking gender into account when examining the genetic profile of patients may help to more specifically identify those who will benefit from 5-FU/oxal chemotherapy.

No significant financial relationships to disclose.

Polymorphisms in angiogenesis related genes predict clinical outcome in patients (pts) with metastatic colorectal cancer (mCRC) treated with first line 5-FU or capecitabine in combination with oxaliplatin and bevacizumab (FOLFOX/BV or XELOX/BV)

10576

Background: The inhibition of angiogenesis is central to the mechanism of action of BV, a monoclonal antibody to vascular endothelial growth factor (VEGF). We evaluated functionally significant polymorphisms of genes involved in the angiogenesis/VEGF pathway as potential molecular predictors of clinical outcome in pts with mCRC who received BV as part of their frontline therapy. These genes included: VEGF, VEGF receptor 2 (KDR), neuropilin 1 (NRP 1), Interleukin (IL) 6 and 8, IL receptor 1 and 2 (CXCR 1, CXCR 2) adrenomedullin (AM), leptin, fibroblast growth factor receptor 4 (FGFR4), tissue factor (TF), matrix metalloproteinases (MMP 2,7,9), epidermal growth factor receptor (EGFR), aryl hydrocarbon receptor nuclear translocator (ARNT), and nuclear factor kappa b (NFkb). Methods: PCR-RFLP assays were performed on genomic DNA extracted from the blood of 30 pts with mCRC treated with first-line FOLFOX/BV or XELOX/BV at USC. Results: the cohort consisted of 21 males and 9 females with a median age of 56 years (range: 29–81). 20 pts received XELOX/BV as part of an on-going phase II study, 10 pts received FOLFOX/BV. Radiologic response was evaluable in 27/30 pts: 2/27 (7%) complete response (CR), 14/27 (52%) partial response (PR), and 10/27 (37%) stable disease (SD) and 1/28 (4%) progressive disease. At a median follow-up of 19.4 months, 16/30 pts progressed with a median progression free survival (PFS) of 11.8 months. Pts homozygous A/A at the leptin 5'UTR region had a higher probability of response than pts with the G/A or G/G genotypes (p=0.03, Fisher's exact test). Pts with one or more G allele (G/G or A/G) at locus -181 in the promoter region of MMP7 had a higher probability of response than pts with the AA genotype (p=0.014, Fisher's exact test). There were statistically significant associations between genomic polymorphisms of KDR, CXCR2, MMP7 and PFS (<0.05, Log-rank test). Conclusions: This pilot study demonstrated, in a preliminary fashion, the potential predictive and prognostic value of several genomic polymorphisms in pts with mCRC treated with FOLFOX/BV or XELOX/BV.

No significant financial relationships to disclose.

S0413: A phase II SWOG study of GW572016 (lapatinib) as first line therapy in patients (pts) with advanced or metastatic gastric cancer

4621

Background: GW572016 is a dual tyrosine kinase inhibitor of EGFR and HER2/ErbB2. Overexpression of EGFR and ErbB2 has been described in 15 to 45% of gastric cancer, making this a potential target in advanced/metastatic tumors. Methods: The primary objective of this study was to assess the response rate (confirmed complete (CR) and partial responses (PR)). Secondary objectives included time to treatment failure (TTF), overall survival (OS), toxicities and the relationship of EGFR, ErbB2, and markers of angiogenesis with clinical outcome in patients treated with GW572016. A two-stage design was used to detect a difference in the null hypothesis of 5% response probability and the alternative 20% response probability. If at least one response occurred after the first 20 pts, another 20 were to be accrued. GW572016 was administered to chemonaiive metastatic gastric cancer patients at a dose of 1,500 mg orally daily. A cycle was defined as 28 days of therapy, and patients were staged after 2 cycles of treatment. Results: The study met its first stage goal, and continued until full accrual. The study accrued 47 pts from February 2005 until May 2006. Two patients lacked required tissue/blood submission but are included in this clinical analysis. One patient did not receive treatment and is not analyzable. Pt characteristics: male/female30/16 (65%/35%); median age 68.7 years (range 38.9 - 90). Significant toxicities: 1 grade 4 cardiac ischemia/infarction, 2 grade 4 fatigue, 1 grade 4 vomiting. There was one treatment related death due to CNS ischemia. Three pts (7%) had a confirmed PR and 2 (5%) unconfirmed PR, and 9 (20%) had stable disease. Median TTF was 2 months, and OS was 5 months. Molecular correlative data was available on 42 pts. Conclusions: GW572016 is a well tolerated regimen with modest single-agent activity in pts with advanced/metastatic gastric cancer. Although, GW572016 did not meet the primary endpoint, this targeted agent may warrant further investigation in combination regimens. Molecular correlatives will be reported in a separate abstract.

No significant financial relationships to disclose.

EGFR, Cox-2, and EGF polymorphisms associated with progression-free survival of EGFR-expressing metastatic colorectal cancer patients treated with single-agent cetuximab (IMCL-0144)

4129

Background: A phase II study of cetuximab (IMCL-0144) has shown response rate of 12 % in patients with EGFR expressing metastatic colorectal cancer (mCRC). Recently, we reported that polymorphisms in the EGFR pathway may be useful molecular markers to predict clinical outcome. In this larger study, we tested whether polymorphisms in genes involved in the EGFR and angiogenesis pathway will be associated with clinical outcome. Methods: We analyzed 136 tissue samples from 346 mCRC pts enrolled in the phase II study of cetuximab (IMCL-0144), 133 cases were informative. The response rate in these 133 pts was 10% with a median progression-free survival (PFS) of 1.3 months (95% CI, 1.2 to 1.5) and an overall survival time (OS) of 5.5 months (95% CI, 4.1 to 7.5). Gr3–4 toxicity was observed in 56%. Gene polymorphisms of EGFR, Cox-2, EGF, cyclin D1, fragment c γ receptor 2A (FCGR2A), FCGR3A, VEGF, IL-8 were assessed from gDNA extracted from tissue samples by using PCR-based RFLP technique. Univariate analysis (Fisher’s exact test for response; log-rank test for PFS and OS) was performed to examine associations between polymorphisms and clinical outcome. A classification and regression tree (CART) analysis was used to identify subgroups of patients who were more likely to benefit from cetuximab. Results: Pts with EGFR G497C GA, Cox-2 G-765C CC, EGF A61G GG genotype showed better PFS (p=0.02, 1.8mo. vs. 1.2mo.; p=0.03, 6.9mo. vs. 1.3mo.; p=0.04, 1.4mo. vs. 1.2mo.), respectively. We found trends in associations between Cox-2 and tumor response (p=0.09), between EGF and Gr3–4 toxixity (p=0.06). CART analyses indicated that germline polymorphisms in EGFR, EGF, Cox-2, Cyclin D1, IL-8, FCGR2A and FCGR3A genes could be used to identify patients who benefit most likely of cetuximab therapy. Conclusions: Our data suggest that the polymorphisms of EGFR, Cox-2, and EGF may be useful molecular markers to predict clinical outcome in mCRC pts treated with single-agent cetuximab. And prospective studies will need to be done to confirm these preliminary findings.

[Table: see text]

VEGF and VEGF receptor-2 (VEGFR2) gene polymorphisms predict tumor recurrence in stage II and III colon cancer

4004

Background: Despite recent advances in the treatment of metastatic colorectal cancer, tailoring adjuvant treatment of stage II and III colon cancer patients remains controversial. Identifying a reliable panel of prognostic and predictive markers for tumor recurrence is critical in selecting an individualized and tailored chemotherapy. Tumor angiogenesis plays an important role in tumor development, progression and metastasis. In this retrospective study, we tested whether a specific pattern of 40 functionally significant polymorphisms in 37 genes involved in angiogenesis and tumor microenvironment will predict the risk of tumor recurrence in stage II and III colon cancer patients treated with adjuvant chemotherapy. Methods: Between 1999 and 2006 blood specimens from 140 patients (69 females and 71 males with a median age of 59 years; range=28–86) were obtained at the University of Southern California/Norris Comprehensive Cancer Center (USC/NCCC). Sixty-three patients had stage II and 77 had stage III colon cancer. The median follow-up was 5.4 years (range=2.0–16.8). 51 of 140 patients (36.4%) developed tumor recurrence with a 5-year probability of 0.28 ± 0.06 for stage II and 0.40 ± 0.06 for stage III colon cancer patients. Genomic DNA was extracted from peripheral blood and genotypes were determined using PCR based RFLP. Results: Polymorphisms in VEGF (C936T; p=0.009, log-rank) and VEGFR2 (+4422 AC- repeat; p=0.04, log-rank and +1416 T/A; p=0.0009, log-rank) were associated with risk of tumor recurrence in stage III colon cancer patients (n=77). VEGFR2 AC-repeat polymorphisms were additionally associated with risk of recurrence in Stage II colon cancer patients (n=63, p=0.02, log-rank). Conclusion: VEGF C936T and VEGFR2 (+4422 AC-repeat and +1416 T/A) polymorphisms may help to identify Stage II and III colon cancer patients who are at increased risk for developing tumor recurrence. Angiogenesis seems to play a crucial role in tumor recurrence, thus targeting VEGF and VEGFR2 may be of clinical benefit for stage II and stage III colon cancer patients. Large prospective trials are needed to validate these preliminary data.

No significant financial relationships to disclose.

Molecular signatures generated with RNA isolated from formalin-fixed paraffin-embedded tumor specimens differentiate metastatic and non-metastatic primary colorectal cancer

3615

Background: Colorectal cancer (CRC) can be either non-metastatic (locally-advanced) or metastatic, with the latter having a considerably worse prognosis. Because tissue from patients is generally available only as FFPE specimens, we investigated whether distinct molecular signatures could be obtained for these CRC types using RNA isolated from archival FFPE specimen blocks in gene expression arrays. Methods: FFPE specimens were available from 12 non-metastatic tumors and 9 primary tumors along with their matching metastases. Up to 5 10-micron sections of each were microdissected to isolate areas of tumor tissue. RNA was extracted using a proprietary procedure of Response Genetics, Inc. and was then amplified and labeled. The resulting cRNA was hybridized to the U133 plus 2.0 GeneChip. Unsupervised PCA analysis of the samples resulted in the first principal component separating 2 distinct groups, which consisted of the non-metastatic and metastatic tumors. A differentially expressed gene list between metastatic and non-metastatic CRC was determined. These data were also analyzed for differential canonical pathways using Ingenuity Pathway Analysis. Results: Hierarchical clustering analysis segregated locally advanced primary tumors and metastatic primary tumors into two clusters with distinct gene signatures. A T-test with unequal variance assumption identified 609 differentially expressed probe sets with FDR (false discovery rate) = 0.05. Comparison of primary tumors with their liver metastases using a paired T-test showed only 2 differentially expressed genes at FDR = 0.05 but 526 genes with significance p value < 0.005. Pathway analysis showed significant deregulation of VEGF, hypoxia, B Cell receptor, PI3K/AKT, ERK/MAPK and G-protein coupled receptor signaling between locally advanced and metastatic tumors. Pathway analysis of primary tumors and metastases showed deregulated TGF-b, integrin and chemokine signaling pathways. Conclusion: We have demonstrated the feasibility of identifying metastatic and non-metastatic tumors by microarray analysis using FFPE CRC tissue. This result is currently being validated in a separate larger cohort of patients.

[Table: see text]

Phase I dose and schedule finding study of pegylated liposomal doxorubicin (D) and weekly docetaxel (T)

12012

Background: The combination of D and T has many potential applications, particularly in breast and ovarian cancers. A phase 3 trial is examining D + T versus T in first-line metastatic breast cancer ( NCT00091442 ). D has better tumor localization and penetration in solid tumors than conventional doxorubicin. In previous studies, the maximum tolerated dose (MTD) of the combination was identified as D 30 mg/m2 and T 75 mg/m2 q4 weeks (wks), with a recommended dose and schedule of D 30 mg/m2 and T 60 mg/m2 q3wks without G-CSF. We conducted a phase I study to determine the MTD of D with weekly T. Our hypothesis was that the lower incidence of myelosuppression with weekly T would allow for higher doses of both drugs. Methods: There were 2 schedules. Arm A: D q4wks starting at 25 mg/m2 with weekly T for 3 wks starting at 30 mg/m2. Arm B: D q2wks starting at 15 mg/m2 with weekly T for 3 wks starting at 30 mg/m2. One cycle was 28 days. Standard 3+3 design was used with MTD defined as the highest dose level causing dose limiting toxicity (DLT) in ≥ 2/6 patients (pts). Results: 32 pts were treated; 13 females, 19 males, median age of 60 years. Median number of cycles administered was 2 (1–13) with a median follow-up of 11.5 months. Tumor types included lung (16%), thyroid (9%), esophagus (9%), nasopharynx, breast, colorectal, stomach and kidney (6% each). Arm A (13 pts) was closed after 2/7 evaluable pts at dose level 2 (D 33mg/m2; T 30 mg/m2) experienced DLT in the form of grade 3 stomatitis. The most common grade 3/4 toxicities were neutropenia (3/13), stomatitis (3/13) and fatigue (3/13). Arm B accrued 19 pts. The trial was closed at the highest planned dose in Arm B (D 20mg/m2 q2wks and T 35 mg/m2 weekly) with only 1/6 evaluable pts experiencing DLT in the form of grade 4 fatigue and weakness. The most common grade 3/4 toxicities in Arm B included neutropenia (5/19 pts), fatigue (5/19 pts) and stomatitis (2/19 pts). There was no grade 3/4 hand-foot syndrome or cardiotoxicity. 2 partial responses were observed in nasopharyngeal and salivary gland carcinomas, with 13 pts achieving stable disease. Conclusions: The combination of D q2 wks and T weekly for 3/4 wks is well tolerated and results in a higher dose intensity of both drugs than in previously evaluated regimens.

[Table: see text]

A randomized phase II non-comparative study of Ispinesib given weekly or every three weeks in metastatic colorectal cancer. A California Cancer Consortium Study (CCC-P)

3595

Background: Ispinesib(SB-715992) is a polycyclic, nitrogen-containing heterocycle that inhibits the mitotic kinesin spindle protein (KSP). KSP is essential for mitotic spindle assembly and function during mitosis, and is a rational target of anti-cancer therapy. This phase II study used two different dosing schedules; the primary objective was to determine the response rate (RR) and the secondary objectives were to determine time to tumor progression (TTP), progression free survival (PFS), overall survival (OS) and toxicity. Methods: Patients (pts) were randomized to receive (Arm A) ispinesib 7 mg/m2 every week for 3 weeks, every 28 days or (Arm B) 18 mg/m2 every 21 days. Response was assessed every 6 weeks. Chemotherapy was administered until disease progression or intolerance. Results: A total of 64 pts were accrued. The median number of cycles was 2 for both arms. Five pts had stable disease and 48 had progressive disease. PFS was 49 days in Arm A (44 to 51) and 37 days in Arm B (35 to 42 days). The most common grade 3/4 toxicities in arms A and B respectively included neutropenia (3 and 20), nausea and vomiting (3 and 1), neurologic (1 and 2). Of these, only 1 pt had febrile neutropenia and 1 pt had peripheral sensory neuropathy. The toxicity data is not available on 2 patients. Eleven pts are not evaluable for response yet. Conclusions: Ispinesib did not demonstrate significant activity in heavily pretreated patients with advanced/metastatic colorectal cancer at the dose and schedule employed in this trial. Correlative studies are in progress. Supported by NO1 CM17101

[Table: see text]

No significant financial relationships to disclose.

Polymorphisms of DNA-repair genes associated with clinical outcome in metastatic breast cancer (MBC) patients treated with gemcitabine/cisplatin (GC) (California Cancer Consortium)

675

Background: DNA repair enzymes may play an important role in determining efficacy of chemotherapy in MBC. In particular, GC combination therapy may be dependent on activity of DNA repair enzymes in host cells, since cisplatin acts by inducing DNA damage. Cancer cells with increased DNA repair capacity may be resistant to GC, and specific genes may be responsible for this increased repair capacity. We examined whether polymorphisms in genes related to DNA repair were associated with clinical outcome in MBC patients treated with GC, enrolled in a parent phase II clinical trial (Ph II-14 A & B). Methods: Fifty-five patients with MBC were evaluated. Patients received the following regimen: 25 mg/m2 cisplatin on days 1–4; 1000 mg/m2 gemcitabine on days 2 and 8 of 21-day cycle. Thirteen polymorphisms in 10 cancer-related genes were tested for association with overall survival, time to tumor progression, and tumor response using a PCR RFLP based assay. Results: Of 55 patients evaluated, there were 17 responders (31%) and 33 non-responders (60%). Five patients (9%) inevaluable for response. Of 33 non-responders, 15 had stable disease, 18 had progressive disease. Median survival: 11.7 months with median follow-up 32.4 months for 4 patients alive at time of analysis. Median progression-free survival: 4.2 months. XPD Lys751Gln polymorphism was associated with overall survival and time to tumor progression (p=0.0003, p=0.006, respectively, log-rank test). Thirty-five patients carried Lys/Lys genotype, of which 29% resopnded. Fourteen patients carried Lys/Gln genotype, of which 54% resopnded. Five patients carried Gln/Gln genotype, with no responders. XRCC3 Thr241Met polymorphism was associated with time to tumor progression and tumor response (p=0.03, p=0.002, respectively). Eighteen patients had Met/Met genotype, of which 47% responded. Twenty-six patients had heterozygous genotype, of which 17% responded. Five patients had homozygous Thr/Thr, of which 100% responded. Conclusions: Our results suggest that polymorphisms in DNA repair genes XPD and XRCC3 may be important markers in predicting clinical outcome in MBC patients treated with GC. Supported by the following NCI grant: N01 CM1701.

[Table: see text]

Polymorphism in sodium-channel alpha 1-subunit (SCN1A) predicts response, TTP, survival, and toxicity in patients with metastatic colorectal cancer treated with 5-FU/oxaliplatin

3533

Background: Genetic polymorphisms in DNA repair and drug metabolism pathways have been shown to be associated with efficacy and toxicity in patients with metastatic colon cancer treated with 5-FU/oxaliplatin. Recent studies demonstrated in in vitro models that the neurotoxicity associated with oxaliplatin may be linked to an effect on neuronal voltage-gated sodium channels (VGSC). In vitro and in vivo models showed that increased VGSC alpha gene expression was associated with metastatic potential, proliferation and progression of breast and prostate cancer indicating a role in predicting toxicity and efficacy to chemotherapy. We tested the hypothesis whether VGSC gene polymorphisms may predict clinical outcome in a phase II study of combination oxaliplatin with 5-FU in patients with colorectal cancer refractory to 5-FU and/or irinotecan based chemotherapy. Methods: 173 patients were enrolled in this phase II study. 152 patients (male/female; 78/74, median age; 60), median follow up of 18.6 months, response rate of 19%, median time to tumor progression 4.2 months and median survival of 10.3 months. Grade 3/4 toxicity was seen in 60% with GI toxicity of 42% and Neurotoxicity of 11%. The dose of oxaliplatin was 130mg/m2 every 3 weeks and 5-FU was 200mg/m2/day CI for 10 weeks followed by 2 weeks rest. We tested the 12 VGSC genes polymorphisms (SCN1A, 1B, 1A1, 1A2, 1A3, 1A4, 1A5, 1A A3169G SNP, 1A C1702T Nonsense Mutation, 1A T1067A SNP, 1A C3637T SNP, SCN8A Ref SNP 303802). Genomic DNA was extracted from peripheral blood samples and polymorphisms were analyzed by PCR-based RFLP technique. Results: Patients with SCN1A T1067A SNP T/T genotype showed a significant better response rate (p=0.02, 21.9% [23/105] vs. 11.3% [5/44]), TTP (p=0.02, 4.6 months vs. 3.4 months), overall survival (p<.001, 12.3 months. vs. 8.0 months.), and frequency of grade 3/4 toxicity (p=.002) compare to patients with T/A genotype. No A/A genotype was observed. Conclusions: SCN1A gene polymorphism may be potential molecular marker for survival and toxicity in patients with colorectal cancer treated with 5-FU/oxaliplatin. In vitro studies are ongoing to identify the mechanism of resistance by SCN1A.

[Table: see text]

Phase II study of lapatinib, a dual inhibitor of epidermal growth factor receptor (EGFR) tyrosine kinase 1 and 2 (Her2/Neu) in patients (pts) with advanced biliary tree cancer (BTC) or hepatocellular cancer (HCC). A California Consortium (CCC-P) Trial

4010

Background: Advanced BTC (gallbladder, bile duct) and HCC respond poorly to systemic chemotherapy. Lapatinib is an oral inhibitor of EGFR and Her/2-neu with evidence of activity in a number of tumor types. Both BTC and HCC overexpress EGFR 1 and 2. In addition, Her2/neu somatic mutations have been reported in HCC, and may predict response to EGFR targeted agents (Bekaii-Saab et al, A 4037, ASCO 2005). This trial conducted by the CCC-P and Univ. Chicago, is designed to determine the efficacy of lapatinib in BTC and HCC pts. Methods: A two-stage design is utilized and pts are stratified by tumor type (BTC or HCC); the study has a 90% power to detect a true response rate ≥20%. Two initial groups of 17 patients for BTC and HCC were accrued; one response from each was required to proceed with accrual to a total of 37 patients in each group. Adequate hematologic, renal, and hepatic function is required. Eligible pts are in Child A group and have ECOG PS of 0–2. One prior treatment regimen is allowed. Lapatinib dose is 1,500 mg/d orally without interruption. One cycle is 28 days and radiological assessment is done every 8 weeks. Results: 49 pts have been accrued (BTC 19 and HCC 30). Pt Characteristics: Male 59%, ECOG PS 0 (18 pts), ECOG PS 1 (22 pts). Median age 66 yrs (range19–82). Median cycles 2 (range 1–12). Grade 3/4 toxicity was noted in 19 pts and included fatigue in 4 pts, elevation of liver enzymes in 4 pts and diarrhea in 2 pts. Nausea, vomiting, rash, anemia and thrombocytopenia were noted in 1 pt each. There was no evidence of cardiac dysfunction. In 17 evaulable pts with BTC, no responses were observed, 5 had stable disease (SD). Among the first 17 pts with HCC, 2 confirmed partial responses have been recorded. In addition 8 pts have SD. The progression free survival is 1.8 mo for both BTC and HCC. Tumor and blood specimens are being analyzed for expression of EGFR, HER-2/neu, status of downstream signal pathway molecules, and correlation to response. Conclusions: Lapatinib is well-tolerated. No activity in BTC was noted and that cohort is closed. Lapatinib activity in HCC is encouraging, and study is close to completion. Source of support: NCI-NO1-CM-57018–16.

[Table: see text]