e20029 Background: The risk of lymph node involvement for cutaneous melanoma increases with tumor thickness - 5% for Breslow thickness ≤1mm and 34% for T4 or ulcerated lesions. SLNB is standard of care for cutaneous melanoma of Breslow thickness >1mm. Positive SLNB is followed by completion lymph node dissection (CLND) of the affected nodal basin. SLNB has a high negative predictive value (NPV) but false negative (FN) rates vary widely (3-12.5%). Methods: Data were collected on 3,600 patients (pts) who underwent SLNB for cutaneous melanoma from 1996 to 2012 using the UPCI Cancer Registry. Cancer registry records were reviewed for primary tumor characteristics, demographic details, and outcomes. The influence of baseline characteristics on SLNB positivity was assessed using Cox proportional hazards analysis. Results: 3,600 SLNB performed from 1996 to 2012 of which 969 SLNB (969 pts) reviewed as a pilot. Primary tumor characteristics and demographic details obtained. SLN metastases were detected in 169 pts (17.4%). 134 CLND were performed with 31 pts not having CLND secondary to refusal or early distant metastatic disease. Following CLND, 48 recurrences were observed: recurrence pattern was predominantly cutaneous (19/48 pts, 39.6%), although ipsilateral recurrences in the prior CLND bed were noted (14/48 pts, 29.2%). Visceral recurrences comprised the remainder [non-pulmonary 7/48 pts (14.6%), pulmonary 5/48 pts (10.4%) and CNS 2/48 pts (4.2%)]. Most CLNDs followed by ipsilateral regional nodal recurrences were performed outside tertiary care centers. PFS and OS were significantly worse with increasing stage and SLNB status. Conclusions: AJCC, ASCO and SSO guidelines recommend SLN biopsy for intermediate-thickness melanomas (1-4mm Breslow) to optimize staging accuracy. Incidence of regional recurrence after CLND has been estimated at 30-35%. In this large and uniformly treated series of patients at a major melanoma referral center, SLN were involved in 17.2% and median PFS after CLND was 22mos. Biomarkers associated with SLNB and CLND positivity are currently being evaluated in prospectively banked specimens from patients (07-133) and under the aegis of the SPORE in Skin Cancer (P50CA121973).

EGFR polymorphisms have not been thoroughly evaluated for association with head and neck squamous cell carcinoma (HNSCC) risk. We genotyped 578 HNSCC patients and 588 cancer-free controls for 60 EGFR single nucleotide polymorphisms (SNPs) and tested associations with HNSCC risk.EGER intronic SNPs rs12535536, rs2075110, rsl 253871, rs845561 and rs6970262 and synonymous SNP rs2072454 were associated with HNSCC risk among all subjects (p < 0.05). SNPs rs12538371, rs845561, and rs6970262 were significantly associated with HNSCC risk (p < 0.05) among never tobacco users. We identified EGFR variants that likely modify risk for HNSCC including three variants that contribute to tobacco-independent risk. (C) 2014 Elsevier Ireland Ltd. All rights reserved.

Abstract Background: EGFR is overexpressed in HNSCC and contributes to disease progression. HNSCC arising in individuals with and without tobacco use histories represent distinct disease etiologies. A thorough evaluation of EGFR polymorphisms for association with HNSCC risk has not been reported. Methods: We determined the genotypes at 60 EGFR single nucleotide polymorphisms (SNPs) in 578 white HNSCC patients and 588 white cancer-free controls using the iPLEX gold mass spectrometry-based genotyping assay. EGFR SNPs were selected based on hypothesized function and/or tagging of EGFR with r2 ≥0.8 and minor allele frequencies of ≥5%. Associations with HNSCC risk were tested using the Fisher's exact test for genotype and for trend for number of alleles of each SNP. We also performed these analyses stratified by tobacco use. SNPs significantly associated with HNSCC (p<0.05 for either test) were further tested using multivariable logistic regression (MLR) analysis. SNPs significant in MLR models were evaluated in an independent cohort of 408 white HNSCC patients and 408 white cancer-free subjects. Expression of non-coding RNA transcript EGFR-AS1 was detected by PCR amplification following reverse-transcription of RNA isolated from Fadu, CAL33 and LICR-LON-HN5 HNSCC cell lines. Results: Of the 60 EGFR SNPs tested, 5 intronic SNPs (rs12535536, rs2075110, rs12538371, rs845561 and rs6970262) and 1 synonymous coding SNP (rs2072454) were found to be associated (p<0.05) with HNSCC risk in the entire case-control cohort after adjustment for sex, age, and use of alcohol and tobacco. Among never tobacco users, rs12538371, rs845561, and rs6970262, residing within introns 15, 20 and 21, respectively, were significantly associated with HNSCC risk (p<0.05) after adjustment for age and sex. Two of these SNPs (rs845561 and rs6970262) reside near histone H3 lysine27 acetylation-enriched regions in ENCODE cell lines and sequences encoding the 2.8 kb spliced non-coding transcript EGFR AS1, which we detected in HNSCC cell lines. rs17586365 was significantly associated with HNSCC risk in MLR models restricted to tobacco users. However, SNPs rs6593206, rs2075110, rs12538371, rs845561 and rs6970262 were not associated with HNSCC risk in an independent cohort where all HNSCC cases had histories of tobacco use. Conclusions: Common EGFR genetic variants may modify risk for HNSCC independent of known risk factors. The difference in tobacco use histories between two different cohorts may account for the discrepancy in the association between certain SNPs and HNSCC risk association. Three EGFR variants may specifically contribute to tobacco-independent HNSCC. The functional consequences of these SNPs or risk-conferring polymorphisms in linkage disequilibrium have yet to be defined. Citation Format: Christopher Fung, Kerry Trent, Sonali Joyce, Pei Zhou, Tomoko Nukui, Daniel E. Weeks, Brenda Diergaarde, Yuanqing Ye, Xifeng Wu, Jian-Min Yuan, Jennifer Grandis, Joel L. Weissfeld, Marjorie Romkes, Ann Marie Egloff. Identification of epidermal growth factor receptor (EGFR) polymorphisms that modify risk for squamous cell carcinoma of the head and neck (HNSCC). [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1340. doi:10.1158/1538-7445.AM2013-1340

9075 Background: IL-2 is a T-cell growth factor tested in a variety of regimens for advanced melanoma (MEL) and renal cell carcinoma (RCC). High-dose IL-2 (600,000-720,000 IU/kg administered intravenously every 8 hours for up to 14 consecutive doses) was approved by FDA for advanced MEL and RCC in 1998 based upon the durability of responses observed. Early studies of HD IL-2 reported overall (OR) and complete response (CR) rates of 16% and 8% respectively. Severe toxicity limited use to specialized centers with standardized protocols, either intensive care (ICU) or oncology specialty settings. The U Pittsburgh has treated 1022 patients with IL-2 at any dosage and we here present outcomes of 550 MEL pts treated with HD IL-2 in an oncology specialty non-ICU setting. Methods: Clinical and radiological data were collected on all pts treated with IL-2 using the UPCI Cancer Registry and Medical Archival System (MARS). Pharmacy records were reviewed for dosing details. The influence of baseline characteristics on treatment outcomes was assessed using Cox proportional hazards analysis. Results: A total of 848 pts received HD IL-2, of which 298 pts had RCC while 550 had MEL. Detailed pharmacy dosing records were reviewed from 176 pts treated over the past 12 years (2000-2012) who received a total of 3738 cycles. Of 165 pts evaluable for response, OR was documented in 24 pts (14.8%) and CR in 5 pts (3.0%). Median overall survival (OS) was 10.0 mos for all patients and 21.5 mos for responders (CR+PR). Median number of doses per cycle was 7. Toxicity was consistent with prior reports. HD IL-2 required ICU transfers in 5% and 1 death was attributed to HD IL-2. Pts with higher baseline lactate dehydrogenase (LDH) had poorer OS (p < 0.05). Conclusions: In this large and uniformly treated series of recent patients treated with IL-2 OR/CR rates with HD IL-2 are 14.8% and 3.0% respectively. Higher LDH is associated with poorer outcome. Biomarkers of response are currently being evaluated in banked clinical specimens collected from patients under the SPORE in Skin Cancer (P50 CA121973).

Background The Specialized Program of Research Excellence (SPORE) in Head and Neck Cancer neoplasm virtual biorepository is a bioinformatics-supported system to incorporate data from various clinical, pathological, and molecular systems into a single architecture based on a set of common data elements (CDEs) that provides semantic and syntactic interoperability of data sets. Results The various components of this annotation tool include the Development of Common Data Elements (CDEs) that are derived from College of American Pathologists (CAP) Checklist and North American Association of Central Cancer Registries (NAACR) standards. The Data Entry Tool is a portable and flexible Oracle-based data entry device, which is an easily mastered web-based tool. The Data Query Tool helps investigators and researchers to search de-identified information within the warehouse/resource through a "point and click" interface, thus enabling only the selected data elements to be essentially copied into a data mart using a multi dimensional model from the warehouse's relational structure. The SPORE Head and Neck Neoplasm Database contains multimodal datasets that are accessible to investigators via an easy to use query tool. The database currently holds 6553 cases and 10607 tumor accessions. Among these, there are 965 metastatic, 4227 primary, 1369 recurrent, and 483 new primary cases. The data disclosure is strictly regulated by user's authorization. Conclusion The SPORE Head and Neck Neoplasm Virtual Biorepository is a robust translational biomedical informatics tool that can facilitate basic science, clinical, and translational research. The Data Query Tool acts as a central source providing a mechanism for researchers to efficiently find clinically annotated datasets and biospecimens that are relevant to their research areas. The tool protects patient privacy by revealing only de-identified data in accordance with regulations and approvals of the IRB and scientific review committee.