The Third Annual PMO Live: A Global Biomarkers Consortium Initiative (formerly known as Global Biomarkers Consortium) will take place in San Francisco, California, on October 31 – November 1, 2014. PMO Live is the only global meeting dedicated to advancing the understanding of value and clinical impact of biomarker research in oncology. Guided by the expertise of leaders in this field, participants will receive a thorough understanding of the current and future landscape of the relevance of tumor biomarkers and how to effectively personalize cancer care in the clinical setting. The following abstracts will be presented at the meeting.
CATEGORY I: Epidemiology and risk factors
Computed Tomography (CT) and Chest X-Ray (CXR) Screening for Lung Cancer:
End Points, Outcomes, and Population-Based Screening
Gary M. Strauss,1,2 Lorenzo Dominioni3
1Tufts University School of Medicine, Boston, Massachusetts, USA; 2Tufts Medical Center, Boston, Massachusetts, USA; 3University of Insubria, Varese, Italy
Background: The National Lung Screening Trial (NLST) generated excitement by demonstrating that CT screening reduces lung cancer mortality in comparison to CXR screening. In randomized population trials (RPTs) on cancer screening, disease-specific mortality is assumed to provide an unbiased measure of screening effectiveness. This is based upon the assumption that randomization produces comparison groups with equal probability of death from the target cancer unless the intervention reduces risk. However, this assumption has been violated in numerous RPTs, leading to uncertainty about the effectiveness of screening for numerous cancers. Objectives of this analysis are to assess whether the mortality end point provided an unbiased measure of screening efficacy in the Mayo Lung Project (MLP) and NLST, the most influential RPTs on CXR and CT screening, respectively. Methods: In MLP, following a normal prevalence CXR, 9192 smokers were randomized to an experimental group (EG) undergoing CXR and cytology every 4 months for 6 years followed by 3 years of observation, or to a control group (CG) observed for 9 years. Controls were advised to undergo annual CXR. In NLST, 53,454 current/former smokers were randomized to EG undergoing annual CT screening or to CG undergoing annual CXR. Participants underwent prevalence screening followed by 2 annual incidence screens. Results: In MLP, non–small cell lung carcinoma (NSCLC) mortality was 13% higher (P=.48), although survival was far superior to EG (P=.0095). NSCLC incidence was 37% higher in EG (P=.0098), leading to the hypothesis that overdiagnosis was responsible for the mortality/survival discrepancy. However, MLP data are inconsistent with overdiagnosis. Randomization failure was responsible for higher NSCLC incidence in EG, confounding the ability of mortality to reflect CXR screening efficacy. Indeed, survival, not mortality, provided an unbiased surrogate for cure in MLP. In NLST, there was a significant 20% lung cancer mortality reduction in EG (P=.0022). There was also a significant increase in lung cancer incidence in this group (P=.0067). Because CT is more sensitive than CXR, higher incidence of lung cancer is predictable based upon lead time and length biases, and possibly overdiagnosis. However, incidence of the most virulent lung cancers, including small cell lung cancer and NSCLC not otherwise specified, was 16% lower in EG (P=.046). These differences suggest imbalances in randomization. Conclusion: In NLST, lower incidence of lethal lung cancer subtypes in EG predicts for lower lung cancer mortality, independent of CT screening efficacy. Accordingly, lower lung cancer mortality in EG does not provide an unbiased measure of CT screening efficacy. Notwithstanding, stage and case fatality differences support that CT is superior to CXR screening. Population-based screening is mandatory to reduce the global burden of lung cancer mortality among the 1.3 billion cigarette smokers on our planet. CXR screening is a viable option in regions where cost is likely to preclude widespread CT screening.
CATEGORY II: Molecular biology and pathogenesis – implications for therapy
Predictive Molecular Analysis of Regional Lymph Node Metastasis in Melanoma: First Steps to Elimination of Sentinel Lymph Node Biopsy
Collins J,1 George K,2 Stafford C,1 Edenfield J,3 Schammel C,4 Horton S,4 Trocha S5
1Medical University of South Carolina, Charleston, South Carolina, USA; 2Furman University, Greenville, South Carolina, USA; 3Institute for Translational Oncologic Research, Greenville Health System, Greenville, South Carolina, USA;
4Pathology Associates, Greenville, South Carolina, USA; 5Greenville Health System, Greenville, South Carolina, USA
The incidence of melanoma worldwide has steadily been increasing for decades. Although clinicopathologic and histopathologic characteristics are generally successful in characterizing and classifying melanoma, molecular biomarkers offer a more conclusive prognosis and the opportunity for individualized treatment. We used exome sequencing to identify molecular markers associated with positive lymph nodes in patients with T2 and T3 melanomas (1-4 mm). Our study population included 9 patients with metastatic melanoma to the lymph nodes, and 9 patients with melanoma localized to the primary tumor site and corresponding SLNBx negative nodes. The 2 cohorts were statistically identical as shown by a one-sided t test comparing the 2 groups on the basis of age (P=.17), race (P=.18), Breslow depth (P=.14), Clark level (P=.33), host response (P=.17), ulceration (P=.50), satellite nodules (P=.17), lymphovascular invasion (P=.50), and mitotic activity (P=.09). Only PIK3CA showed a statistically significant difference (P=.03) between the 2 cohorts. Further analysis using classification and regression tree (CART) analysis was conducted. Of the patients that were PIK3CA positive (n=6), 5 were SLN positive (83%); furthermore, a lack of mutation in the NRAS gene guaranteed that our PIK3CA-positive patients would end up as SLN-positive patients (100%). Interestingly, the only patient that was PIK3CA positive and SLN negative was also NRAS positive (100%). For the rest of our cohort that was PIK3CA negative (n=14), all the patients with a mutation in the ERBB4 gene (n=3) were SLN negative. Of patients that were PIK3CA and ERBB4 negative (n=11), 7 were TP53 negative (63.3%). Of these patients, there was a 71.5% chance that they would be SLN negative. The other 4 patients as yet unaccounted for in the cohort were TP53 positive. Of those 4, only 1 was SMAD4 positive and was SLN negative. Those patients that were TP53 positive and SMAD4 negative (n=3) were all SLN positive. In the emerging world of personalized medicine utilizing molecular evaluation and now in combination with powerful statistical analyses, we demonstrate an exciting first step in the new era of evaluation of metastatic potential in melanoma. While evaluation of lymph node metastases in melanoma, shown to be the most important prognostic indicator, is typically dominated by surgical excision of a sentinel node, utilizing molecular analyses, it may be possible to apply surgery only to those who will benefit from the risk. While this pilot study shows the possibility of creating a predictive model for metastatic potential, a larger patient population is necessary to generalize the model.
CATEGORY VII: Emerging translational data – impact on future therapy
Next-Generation Sequencing Demonstrates Association Between Tumor Suppressor Gene Aberrations and Poor Outcome in Patients With Cancer
Maria Schwaederle,1 Gregory A. Daniels,1 David E. Piccioni,1 Paul T. Fanta,1 Richard B. Schwab,1 Kelly A.
Shimabukuro,1 Barbara A. Parker,1 Razelle Kurzrock1
1Center for Personalized Cancer Therapy, and Division of Hematology and Oncology, UCSD Moores Cancer Center, La Jolla, California, USA
Background and Objectives: Next-generation sequencing is advancing at a breathtaking pace and is transforming patient care by allowing physicians to customize and match treatment to their patients’ tumor alterations. Genomic aberrations may be important because they predict response to therapy or because they are prognostic. Our primary objective was to study the association between key molecular alterations and outcome parameters. Methods: This was a retrospective, correlative and exploratory study that integrated clinicopathology and clinical outcomes of 392 patients seen at the UC San Diego Moores Cancer Center (October 2012 to April 2014) that had clinical grade molecular testing performed (next-generation sequencing of 236 cancer-related genes). Potential associations between clinical characteristics and outcome parameters (overall survival [OS]), time to metastasis/
recurrence, and best progression-free survival (PFS) were examined. The Kaplan-Meier method and Cox regression models were used for our analysis, and results were subjected to internal validation using a resampling method (bootstrap analysis). Results: In a multivariable analysis (Cox regression model), the parameters that were statistically associated with a poorer OS were the presence of metastasis at diagnosis (P=.014), gastrointestinal histology (P<.0001), PTEN (P<.0001), and CDKN2A alterations (P=.0001). The variables associated with a shorter time to metastasis/recurrence were gastrointestinal histology (P=.004), APC (P=.008), PTEN (P=.026), and TP53 (P=.044) alterations. The analysis of best PFS demonstrated that, in our population, TP53 (P=.003) and PTEN (P=.034) alterations were independent predictors of a shorter best PFS; the type of treatment was also associated with best PFS (P=.039) (with hormonal therapy having a better outcome), as was a personalized treatment approach (matching at least one abnormality in a patient molecular profile to a cognate targeted drug) (P=.046). Conclusion: Our study demonstrated that anomalies in several tumor suppressor genes, including TP53, PTEN, CDKN2A, and APC had a negative prognostic association in our patients.
Cyclin Alterations in Diverse Cancers: Outcome and Coamplification Network
Maria Schwaederle,1 Gregory A. Daniels,1 David E. Piccioni,1 Paul T. Fanta,1 Richard B. Schwab,1 Kelly A.
Shimabukuro,1 Barbara A. Parker,1 Razelle Kurzrock1
1Center for Personalized Cancer Therapy, and Division of Hematology and Oncology, UCSD Moores Cancer Center,
La Jolla, California, USA
Background and Objectives: Cyclin genes are key regulatory components of the cell cycle. Importantly, there are several potent targeted agents in development that can impact them. Our primary objective was to examine the biological implications and clinical and outcome associations of cyclin gene (CCN) alterations in diverse malignancies. Methods: This was a correlative study that integrated clinicopathology and clinical outcomes of 392 patients seen at the UC San Diego Moores Cancer Center (October 2012 to April 2014) that had clinical-grade, next-generation sequencing performed (about 236 cancer-related genes were assessed). Clinical characteristics as well as outcome parameters (best progression-free survival [PFS]), time to metastasis/recurrence, metastatic sites, and overall survival (OS) were compared for patients harboring cyclin alterations versus not. Correlations with other molecular alterations were also studied. Results: CCN alterations were found in 13% of our population (50/392; all were amplifications) and were associated with breast cancer (P<.0001) and a higher median number of alterations (median, 8 vs 3 alterations, P<.0001). A multiple logistic regression analysis showed that patients with CCN amplifications had more liver metastases (P=.046). Harboring a cyclin amplification was not associated with OS, time to metastasis/recurrence, nor with the best PFS. In a Cox regression model, gastrointestinal histology, PTEN, and CDK alterations had a significant association with poorer overall survival. CCN amplifications correlated with alterations in FGF/FGFR family genes as well as in MET and ARFRP1 (all P<.05). An extended correlation study shed light on a network of coamplifications influenced in part by genes that were localized on the same amplicons. Conclusion: CCN amplifications are common across cancers and had distinctive biological associations: correlation with higher numbers of aberrations, and with liver metastases, as well as with abnormalities in FGF/FGFR, MET, and ARFRP1 genes. Customized combinations targeting the cyclin pathway as well as the extended coamplification network may be necessary in order to address resistance mechanisms.