Dueck, Amylou C.

LBA4 Background: Lapatinib (L) is a HER1-HER2 tyrosine kinase inhibitor. The Adjuvant Lapatinib and/or Trastuzumab Treatment Optimisation (ALTTO) Trial is a randomised, phase III trial comparing 3 oral L-containing regimens with T, each given for 1 year. Methods: From June 2007 to July 2011, 8381 patients (pts) were randomised from 946 sites in 44 countries to receive either L+T, T→L, L, or T. Anti-HER2 therapy was initiated after completing all chemotherapy (N=4613), concurrently with a taxane following anthracycline (N=3337), or concurrently with a non-anthracycline, platinum-containing regimen (N=431). The L arm was closed in Aug 2011 for futility and is not presented. The primary endpoint is invasive disease-free survival (DFS). L+T vs. T is tested for superiority and T→L vs. T is tested for non-inferiority at 1.11 margin. P≤0.025 is required for statistical significance. 850 DFS events in the L+T vs. T comparison would provide 80% power to detect a true hazard ratio (HR) of 0.80 with experiment-wide alpha=0.05. The current analysis was planned to occur either after observing these 850 events or at 4.5 yrs median follow up (MFU). Results: Pt and disease characteristics were well balanced. 40% were node-negative and 57% were hormone receptor positive. Only 555 DFS events for the L+T vs. T comparison were observed at 4.5 years MFU. HR for DFS was 0.84 (97.5% CI, 0.70-1.02; P=0.048; 4-yr DFS%=88% vs. 86%) for L+T vs. T and 0.93 (97.5% CI, 0.76-1.13; non-inferiority P=0.044; 4-yr DFS%=87% vs. 86%) for T→L vs. T. Diarrhoea (75% vs. 20%), rash (55% vs. 20%) and hepatobiliary (23% vs. 16%) adverse events were more frequent in L+T vs. T. Primary cardiac endpoints were infrequent (<1%) in all arms. Conclusions: L+T has lower risk of a DFS event compared with T, and T→L appeared non-inferior to T, but neither finding was statistically significant. The first DFS results of dual HER2 blockade in the adjuvant ALTTO at 4.5 years MFU are unexpected considering the effect shown by doubling the pCR rate with L+T vs. T in the NeoALLTO trial. Follow up continues. Clinical trial information: NCT00490139.

137 Background: Diabetes has been associated with an increased risk of mortality among patients with many types of cancers. Methods: We performed a case-control study of 109 breast cancer patients to determine the impact of diabetes on disease recurrence and survival. These 109 patients diagnosed with invasive breast cancer during 2007-2011 with diabetes were identified from the institutional Cancer Registry and matched to 109 breast cancer patients without diabetes according to age, race/ethnicity and year of diagnosis. Statistical tests included paired t-tests, McNemar’s tests, and stratified Cox regression. Results: The median ages of those with and without diabetes were 68 years, and 91% in both groups were white (both matched). Patients with and without diabetes were not significantly different in terms of tumor grade, cancer stage, estrogen or progesterone receptor status, HER2 neu status, marital and employment status, smoking use, use of adjuvant chemotherapy, and steroid treatment (all P >.4). Patients with diabetes reported less use of alcohol (50% vs 64%, P=.04), and had greater BMI (79% overweight/obese vs 58%, P= .002). Survival analysis demonstrated no differences in survival from initial breast cancer diagnosis (HR=1.25, 95% CI 0.49-3.17) with median follow-up of 2.2 years (range 0.1-4.9), and no differences in recurrence-free survival (HR=1.00, 95% CI 0.14-7.10) though very few deaths and recurrences were observed. Within the diabetes group, 96 (88%) had type 2 diabetes and 97/106 (92%, 3 unknown) had diabetes pre-existing to cancer diagnosis. Initial diabetes therapy included diet (33, 30%), oral therapy (48, 44%), insulin (21, 19%), or oral+insulin (6, 6%). 7 patients required a change in diabetes therapy within 1 year of cancer diagnosis with 3 patients starting insulin. 32/70 (46%) with available lab data had 1 or more A1C >/=7 within 1 year of cancer diagnosis. Conclusions: In this retrospective, carefully matched case-control study, we developed a comprehensive profile of the patient population with newly diagnosed breast cancer and diabetes. Diabetes did not appear to impact mortality or recurrence-free survival based on descriptive estimates in this small breast cancer cohort.

13 Background: A Memorial Sloan Kettering Cancer Center (MSKCC) nomogram predicts disease−specific survival (DSS) following R0 resection for gastric adenocarcinoma using patient and tumor factors, with predictive ability superior to that of the current AJCC staging system. However, validation has only been performed in high−volume, specialized institutions. The goal of this study is to compare nomogram-predicted survival with actual survival by using a population database with the largest sample size reported to date. Methods: A retrospective review of all gastric cancer patients from Surveillance, Epidemiology and End Results (SEER) registry data from 1988 to 2000 was conducted. Data for patient age, tumor size, gender, primary location of tumor, Lauren histology, depth of tumor invasion, and % (+) and (-) lymph nodes was entered into the nomogram calculator for each patient. Nomogram-derived 5−year DSS was compared with actual DSS survival using Cox proportional hazards regression and Kaplan-Meier estimates. Results: 4,700 patients were identified. Mean patient age was 68.7 years (SD 12.3), and mean tumor size was 4.8 cm (SD 2.7). The commonest tumor location was the body/middle third of stomach (1519, 32%), followed by the gastro−esophageal junction (1381, 29%) and the antrum (1376, 29%). The majority of patients (76%) had intestinal histology. The estimated proportion of patients surviving 5 years was 23%, 48%, 58%, and 82% for survival groups defined by nomogram likelihoods of 5−year DSS of 0−25%, 26−50%, 51−75%, and 76−100%, respectively. Relative to patients in the 76−100% 5−year nomogram group, patients in the 0−25%, 26−50%, and 51−75% group had significantly higher likelihood of death with hazard ratios of 7.0 (95% CI 6.1−8.0), 3.3 (95% CI 2.8−4.0), and 2.6 (95% CI 2.2−3.0), respectively (all p<0.001). Median follow-up was 9.1 years (range, 0−18). The concordance index (c−index) for the 5−year nomogram score was 0.68 (95% CI 0.67−0.69). Conclusions: There is good discrimination of survival between nomogram-predicted survival groups. Our results validate the use of the MSKCC nomogram in the setting of the general population, outside of specialized institutions, for predicting DSS following resection of gastric adenocarcinoma.

AbstractPatients with indolent non‐Hodgkin lymphoma (NHL) have multiple treatment options yet there is no consensus as to the best initial therapy. Lenalidomide, an immunomodulatory agent, has single agent activity in relapsed lymphoma. This trial was conducted to assess feasibility, efficacy, and safety of adding lenalidomide to rituximab, cyclophosphamide, and dexamethasone (LR‐CD) in untreated indolent NHL patients requiring therapy. This was a single institution phase II trial. Treatment consisted of IV rituximab 375 mg/m2 day 1; oral lenalidomide 20 mg days 1–21; cyclophosphamide 250 mg/m2 days 1, 8, and 15; and dexamethasone 40 mg days 1, 8, 15, and 22 of a 28‐day cycle. Treatment continued 2 cycles beyond best response for a maximum of 12 cycles without rituximab maintenance. Thirty‐three patients were treated. Median age was 68 (43–83 years). 39% had stage IV disease. Histologic subtypes included 8 follicular lymphoma (FL), 7 marginal zone lymphoma (MZL) (1 splenic, 2 extranodal, and 4 nodal), 15 Waldenström's macroglobulinemia (WM), 1 lymphoplasmacytic lymphoma, 1 small lymphocytic lymphoma, and 1 low‐grade B‐cell lymphoma with plasmacytic differentiation (unable to be classified better as MZL or LPL). Hematologic toxicity was the most common adverse event. Median time of follow‐up was 23.4 months (range 1.8–50.9). The overall response rate was 87.9%, with 30.3% complete response. The median duration of response was 38.7 months. The median progression free survival was 39.7 months, while median overall survival (OS) has not yet been reached. Lenalidomide can be safely added to a simple regimen of rituximab, oral cyclophosphamide, and dexamethasone and is an effective combination as initial therapy for low‐grade B‐cell NHL.

INTRODUCTION Acellular dermal matrices have been used with increasing frequency in both reconstructive and cosmetic surgery. While many studies have described the safety and morbidity profiles of these materials, little is known about the relative mechanical properties of individual sheets of allograft harvested from distinct donors. METHODS Sixty-two individual sheets of an acellular dermal matrix from distinct lot numbers (signifying different donors of the dermis) were prospectively analyzed before use. Distribution of thickness according to manufacturer specifications in the dry state were as follows: 0.009 inches to 0.013 inches (1 [1.6%]); 0.79 mm to 1.78 mm (3 [4.8%]); 0.79 mm to 2.03 mm (5 [8%]); 0.8 mm to 3.3 mm (1 [1.6%]); 1.8 mm to 3.3 mm (10 [16.1%]) and 28 mm (6 [9.7%]). The size of the matrix was recorded while dry, after hydration and following stretch. The percentage change in surface area was recorded for each lot. RESULTS The 62 reconstructive cases included breast implant reconstruction (2 [3.2%]); ventral hernia repair (11 [17.7%]); abdominal closure following autologous tissue harvest (6 [9.6%]); autologous breast reconstruction (37 [59.6%]); extremity wound closures (3 [4.8%]) and reinforcement of vertical rectus abdominis muscle closure (3 [4.8%]). The mean percentage change in the size of the acellular dermal matrix to the hydrated state was 58% (36 of 62; thickness 0.06 mm to 3.30 mm); the mean percentage change in size from dry state was 7.14% (range 0% to 18.7%). The mean percentage change in the size of the hydrated matrix to the stretched state was 25.7% (range 0.25% to 70.6%). The variability in elasticity among the individual sheets was significant (P<0.0005). CONCLUSION The acellular dermal matrix displayed highly variable elastic properties among distinct donors. This may be significant in procedures in which symmetry is critical.

AbstractBackgroundGlioblastoma (GBM) exhibits profound intratumoral genetic heterogeneity. Each tumor comprises multiple genetically distinct clonal populations with different therapeutic sensitivities. This has implications for targeted therapy and genetically informed paradigms. Contrast-enhanced (CE)-MRI and conventional sampling techniques have failed to resolve this heterogeneity, particularly for nonenhancing tumor populations. This study explores the feasibility of using multiparametric MRI and texture analysis to characterize regional genetic heterogeneity throughout MRI-enhancing and nonenhancing tumor segments.MethodsWe collected multiple image-guided biopsies from primary GBM patients throughout regions of enhancement (ENH) and nonenhancing parenchyma (so called brain-around-tumor, [BAT]). For each biopsy, we analyzed DNA copy number variants for core GBM driver genes reported by The Cancer Genome Atlas. We co-registered biopsy locations with MRI and texture maps to correlate regional genetic status with spatially matched imaging measurements. We also built multivariate predictive decision-tree models for each GBM driver gene and validated accuracies using leave-one-out-cross-validation (LOOCV).ResultsWe collected 48 biopsies (13 tumors) and identified significant imaging correlations (univariate analysis) for 6 driver genes: EGFR, PDGFRA, PTEN, CDKN2A, RB1, and TP53. Predictive model accuracies (on LOOCV) varied by driver gene of interest. Highest accuracies were observed for PDGFRA (77.1%), EGFR (75%), CDKN2A (87.5%), and RB1 (87.5%), while lowest accuracy was observed in TP53 (37.5%). Models for 4 driver genes (EGFR, RB1, CDKN2A, and PTEN) showed higher accuracy in BAT samples (n = 16) compared with those from ENH segments (n = 32).ConclusionMRI and texture analysis can help characterize regional genetic heterogeneity, which offers potential diagnostic value under the paradigm of individualized oncology.