Citations

Abstract Survival analysis is a widely used method to establish a connection between a time to event outcome and a set of potential covariates. Accurately predicting the time of an event of interest is of primary importance in survival analysis. Many different algorithms have been proposed for survival prediction. However, for a given prediction problem it is rarely, if ever, possible to know in advance which algorithm will perform the best. In this paper we propose two algorithms for constructing super learners in survival data prediction where the individual algorithms are based on proportional hazards. A super learner is a flexible approach to statistical learning that finds the best weighted ensemble of the individual algorithms. Finding the optimal combination of the individual algorithms through minimizing cross-validated risk controls for over-fitting of the final ensemble learner. Candidate algorithms may range from a basic Cox model to tree-based machine learning algorithms, assuming all candidate algorithms are based on the proportional hazards framework. The ensemble weights are estimated by minimizing the cross-validated negative log partial likelihood. We compare the performance of the proposed super learners with existing models through extensive simulation studies. In all simulation scenarios, the proposed super learners are either the best fit or near the best fit. The performances of the newly proposed algorithms are also demonstrated with clinical data examples.

Abstract Background The germline cancer predisposition genes associated with increased risk of each clinical subtype of breast cancer, defined by estrogen receptor (ER), progesterone receptor (PR), and HER2, are not well defined. Methods A total of 54 555 invasive breast cancer patients with 56 480 breast tumors were subjected to clinical hereditary cancer multigene panel testing. Heterogeneity for predisposition genes across clinical breast cancer subtypes was assessed by comparing mutation frequencies by gene among tumor subtypes and by association studies between each tumor subtype and reference controls. Results Mutations in 15 cancer predisposition genes were detected in 8.6% of patients with ER+/HER2-; 8.9% with ER+/HER2+; 7.7% with ER-/HER2+; and 14.4% of ER-/PR-/HER2- tumors. BRCA1, BRCA2, BARD1, and PALB2 mutations were enriched in ER- and HER2- tumors; RAD51C and RAD51D mutations were enriched in ER- tumors only; TP53 mutations were enriched in HER2+ tumors, and ATM and CHEK2 mutations were enriched in both ER+ and/or HER2+ tumors. All genes were associated with moderate (odds ratio > 2.00) or strong (odds ratio > 5.00) risks of at least one subtype of breast cancer in case-control analyses. Mutations in ATM, BARD1, BRCA1, BRCA2, CHEK2, PALB2, RAD51C, RAD51D, and TP53 had predicted lifetime absolute risks of at least 20.0% for breast cancer. Conclusions Germline mutations in hereditary cancer panel genes confer subtype-specific risks of breast cancer. Combined tumor subtype, age at breast cancer diagnosis, and family history of breast and/or ovarian cancer information provides refined categorical estimates of mutation prevalence for women considering genetic testing.

78 Background: Previous study demonstrated that activation of RAS/MAPK pathway is associated with reduced tumor infiltrating lymphocytes and poor response to neoadjuvant chemotherapy in triple negative breast cancer (TNBC). Further study showed that inhibition of MAPK pathway with a MEK inhibitor is synergistic with anti-PD1/PD-L1 therapies. Methods: Patients with unresectable locally advanced or metastatic TNBC with ≤ 3 prior lines of therapy without prior anti-PD-1/PD-L1/PD-L2 therapies were enrolled. Treatment was started with a 2-week run in period with single agent binimetinib. Dose level 0 was binimetinib at 45 mg oral twice daily continuously and dose level -1 was 30 mg twice daily. Pembrolizumab was given at a fixed dose of 200 mg every 3 weeks at both dose levels. Phase I study was based on the standard 3+3 design. Results: A total of 12 patients were enrolled and treated in the phase 1. Five patients were enrolled at dose level 0, 1 patient withdrew prior to treatment and 1 patient was not evaluable for dose limiting toxicity (DLT). Among 3 evaluable patients, 2 patients experienced DLT with grade 3 flank pain and ALT abnormality. Additional 8 patients were enrolled at dose level -1. Out of 6 evaluable patients for DLT, there was 1 DLT observed with grade 3 AST and ALT abnormality. However, this particular patient had liver metastasis with grade 1 AST and ALT abnormality at baseline and her liver function test (LFT) normalized in 3 weeks after treatment discontinuation and oral prednisone. Other grade 1-2 common AEs included rash, LFT increase, abdominal pain, mucositis, nausea, cardiac troponin T increase without EKG change. The efficacy data will be presented at the meeting after the phase II interim analysis. Conclusions: Pembrolizumab in combination with binimetinib at 30 mg twice daily appears to be safe based on the initial cohort. Phase II part is currently ongoing with binimetinib 30 mg twice daily and pembrolizumab 200 mg every 3 weeks with a total of 23 patients planned where the safety and efficacy of this combination will be further evaluated. Clinical trial information: NCT03106415.

PURPOSE In studies of men of European ancestry, rare pathogenic variants in DNA repair pathway genes have been shown to be associated with risk of aggressive prostate cancer. The contribution of rare coding variation to prostate cancer risk in men of African ancestry has not been established. METHODS We sequenced a panel of 19 DNA repair and cancer predisposition genes in 2,453 African American and 1,151 Ugandan cases and controls with prostate cancer. Rare variants were classified as pathogenic or putatively functionally disruptive and examined in association with prostate cancer risk and disease aggressiveness in gene and pathway-level association analyses. RESULTS Pathogenic variants were found in 75 of 2,098 cases (3.6%) and 31 of 1,481 controls (2.1%; odds ratio [OR], 1.82; 95% CI, 1.19 to 2.79; P = .0044), with the association being stronger for more aggressive disease phenotypes (OR, 3.10; 95% CI, 1.54 to 6.23; P = .0022). The highest risks for aggressive disease were observed with pathogenic variants in the ATM, BRCA2, PALB2, and NBN genes, with ORs ranging from approximately 4 to 15 in the combined study sample of African American and Ugandan men. Rare, nonpathogenic, nonsynonymous variants did not have a major impact on risk of overall prostate cancer or disease aggressiveness. CONCLUSION Rare pathogenic variants in DNA repair genes have appreciable effects on risk of aggressive prostate cancer in men of African ancestry. These findings have potential implications for panel testing and risk stratification in this high-risk population.

BACKGROUND: Historically, one in five Hodgkin Lymphoma (HL) patients treated with bleomycin develop bleomycin pulmonary toxicity (BPT), a life-threatening interstitial pneumonitis. The current treatment paradigm consists of prompt discontinuation of bleomycin, administration of corticosteroids, antibiotics, hospital admission, respiratory therapy, or intensive care. BPT represents a challenge for patients with HL as it not only impairs respiratory function, but also has negative impacts on clinical outcomes. A collection of a dozen studies suggest BPT has a mortality rate just short of 10% of patients who develop BPT. Given recent data on bleomycin omission with negative interim PET scan, we assessed changes in BPT rates and severity over the past 15 years. Overall, treatment protocol characterization and patient characteristics most responsive to BPT treatment are poorly understood. OBJECTIVES: In this study, we investigated the clinical impact and treatment strategies in patients with BPT. Our goals were to: 1) Identify HL patients in the last decades who developed BPT and identify risk factors for BPT, 2) evaluate the impact of BPT on long-term clinical outcomes, and 3) characterize patterns of treatment strategies among patients with HL who develop BPT. METHODS: A single-center, retrospective analysis was preformed using patient data from the Mayo Clinic Lymphoma Database (Rochester, MN) consisting of 1,299 patients diagnosed with HL. All patients were diagnosed between 2003-2018. Inclusion criteria included 1) newly diagnosed, biopsy proven HL, 2) upfront treatment with ABVD, 3) treatment was received at our institution. All patients were assessed for clinically relevant HL characteristics including stage of disease, presence of bulky disease, presence of B symptoms, Eastern Cooperative Oncology Group (ECOG) performance status, BPT risk factors, and bleomycin treatment regimen. BPT was clinically defined as 1) Presence of pulmonary symptoms, 2) bilateral interstitial infiltrates on imaging, and 3) no evidence of infectious etiology. Patients treated in the "Early Era" (2000s) were compared to patients in the "Recent Era" (2010s). Comparison of continuous variables between BPT groups was assessed with Wilcox rank-sum test. OS and PFS were estimated via the Kaplan-Meier method. Approval of the protocol by the Mayo Clinic Institutional Review Board (IRB) was obtained and all patients were consented accordingly. RESULTS: One-hundred twenty six patients met the inclusion criteria for this study. Median follow-up for PFS and OS was 5.5 years (95%CI = 4.8-6.4) and 5.8 years (95%CI = 5.0-7.0), respectively. The 10-year OS and PFS among all patients were 85.1% (95%CI = 77.8-93.1) and 86.3% (95%CI = 80.1-93.0), respectively. Forty-seven patients (37% of all patients) met criteria for BPT. The estimated 10-year OS for BPT and non-BPT patients were 74.7% (95%CI = 61.8-90.5) and 91.7% (95%CI = 83.9-100.0), respectively. The estimated 10-year PFS for BPT and non-BPT patients were 84.7% (95%CI = 74.8-95.8) and 87.0% (95%CI = 79.1-95.8), respectively. In univariable analysis, BPT negatively impacted OS (HR=3.6, 95%CI: 1.2-10.6). However, bleomycin omission did not impact OS (HR=1.3, 95%CI=0.5-3.7). BPT-mortality was 17%. In multivariable analysis, BPT was not significantly associated with OS after adjusting for baseline characteristics (HR=3.0, 95%CI=0.9-9.9). Patients were older (median: 46 vs 33 years) and received less bleomycin (median: 107 vs 215 units) compared to non-BPT patients. BPT was most often managed with bleomycin omission with 59% of patients (74 of 126 patients) having omitted bleomycin at some point during treatment. Patients treated in the "Recent Era" (2010s) had lower BPT rates (28% vs 48%), mortality (10% vs 21%), bleomycin dose (143 vs 204 units), and bleomycin cycles (7 vs 12 cycles), yet higher prophylactic bleomycin omission (59% vs 8%) compared to "Early Era" (2000s). Patients treated in the Recent Era compared to Early Era had a reduction of BPT treatment with steroids, hospital admission, respiratory therapy, and ICU admission by 12%, 22%, 14%, 13%, respectively. CONCLUSION: Overall, our data suggests BPT continues to impact OS in HL patients treated with ABVD, however BPT treatment is decreasing as management changed in recent years. Disclosures Ansell: Bristol-Myers Squibb: Research Funding; Mayo Clinic Rochester: Employment; LAM Therapeutics: Research Funding; Trillium: Research Funding; Affimed: Research Funding; Seattle Genetics: Research Funding; Affimed: Research Funding; Mayo Clinic Rochester: Employment; Regeneron: Research Funding; Trillium: Research Funding; Seattle Genetics: Research Funding; LAM Therapeutics: Research Funding; Regeneron: Research Funding; Bristol-Myers Squibb: Research Funding.

Purpose: Multiple myeloma (MM) is a plasma cell (PC) malignancy with an increasing incidence in the US. Epidemiological studies demonstrate a 2-3 fold higher incidence of the pre-malignant monoclonal gammopathy of undetermined significance (MGUS) and MM with a ~4-year younger age of onset among African Americans (AA) compared to European Americans (EAs) (Fonseca, Leukemia, 2017). With equal access to care, AAs have better overall survival compared to EAs (Waxman, Blood, 2010). This disparity may be explained by ancestral-associated genetic predisposition of AAs to development of monoclonal gammopathies and to specific acquired, cytogenetically-defined subtypes. Using calculated genetic ancestry data, we have previously identified a higher prevalence of IgH translocations t(11;14), t(14;16) and t(14;20) in individuals with >80% African ancestry (Baughn, BCJ, 2018). Since SNP rs9344 encoding the CCND1 870G>A polymorphism has been reported in association with increased risk of t(11;14) (Weinhold, Nature Genetics, 2013), we investigated whether rs9344 correlates with African ancestry and with t(11;14) in our cohort of patients with plasma cell dyscrasias. Methods: We studied 898 patients with monoclonal gammopathies who had undergone uniform testing to identify MM-specific cytogenetic abnormalities. DNA from bone marrow samples was genotyped on the Precision Medicine Research Array and biogeographical ancestry was assessed using the Geographic Population Structure Origins tool (Elhaik, Nat Commun, 2014). Plasma cell proliferative disorder FISH of immunoglobulin (cIg)-stained positive plasma cells was performed as described (Baughn, BCJ, 2018). Individuals were divided into three ancestral groups: 1. EAs (<0.1% African ancestry and <30% Asian ancestry); 2. AAs (> 80% African ancestry); and 3. Other. Chi-squared test was used to determine the overall comparison between the 3 ancestral groups and also between ancestral groups 1 vs 2, and 2 vs 3 using pairwise comparison. All tests were two-sided with alpha level set at 0.05 for overall statistical significance. Pairwise comparison was considered statistically significant when p<0.025 based on Bonferroni method for multiple comparisons. Results: We identify increased risk of development of either a t(11;14), t(14;16) or t(14;20) in AAs (48.8%) compared to EAs (33.6%) (p-value=0.0051). To explore the genetic basis of increased t(11;14) specifically (37.4% AAs vs. 27.3% p-value=0.049), we evaluated the frequency of the G risk allele of rs9344 in relation to African ancestry. The frequency of the G risk allele was higher in AAs (0.81) compared to EAs (0.59) (p-value <0.0001) and also higher in t(11;14) cases (0.73) compared to non-t(11;14) controls (0.58) (p-value <0.0001). A multivariate model identified only rs9344 as significantly associated with t(11;14) after adjusting for age, gender and race group suggesting that it plays a role in the development of t(11;14) (p-value <0.001 for GG, p-value=0.005 for AG). To test if these results are replicable, we studied the MMRF CoMMpass cohort. This cohort includes individuals with newly diagnosed MM along with self-report race information and translocation data from long insert whole genome sequencing. Although t(11;14) was not enriched in self-reported Black individuals from this cohort (17.5% Black vs. 20.4% White, p-value=0.47), there was association between the rs9344 and self-reported Black race (p-value <0.0001) and with the presence of t(11;14) when the t(11;14) was analyzed by long insert whole genome sequencing (p-value=0.0001). Conclusions: To our knowledge, this study includes the largest group of African individuals with an abnormal plasma cell clone along with uniformly collected FISH, genotyping and ancestry data. We have identified in this diverse population the association of the CCND1 870G>A polymorphism (rs9344) with African ancestry and with t(11;14) suggesting that it plays a role in the development of t(11;14) plasma cell dyscrasias. Disclosures Elhaik: DNA Diagnostics Center: Consultancy. Baird:DNA Diagnostics Center: Employment. Cerhan:Janssen: Membership on an entity's Board of Directors or advisory committees; Celgene: Research Funding; NanoString: Research Funding. Bergsagel:Celgene: Consultancy; Ionis Pharmaceuticals: Consultancy; Janssen Pharmaceuticals: Consultancy. Dispenzieri:Alnylam: Research Funding; Akcea: Consultancy; Intellia: Consultancy; Janssen: Consultancy; Pfizer: Research Funding; Takeda: Research Funding; Celgene: Research Funding. Kumar:Janssen: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Takeda: Research Funding.

Applications in early-phase cancer trials have motivated the development of many statistical designs since the late 1980s, including dose-finding methods, futility screening, treatment selection, and early stopping rules. These methods are often proposed to address the conventional cytotoxic therapeutics for neoplastic diseases and cancer. Recent advances in precision medicine have motivated novel trial designs, most notably the idea of master protocol (eg, platform trial, basket trial, umbrella trial, N-of-1 trial), for the evaluation of molecularly targeted cancer therapies. In this article, we review the concepts and methodology of early-phase cancer trial designs with a focus on dose finding and treatment screening and put these methods in the context of platform trials of molecularly targeted cancer therapies. Because most cancer trial designs have been developed for cytotoxic agents, we will discuss how these time-tested design principles hold relevance for targeted cancer therapies, and we will delineate how a master protocol may serve as an efficient platform for safety and efficacy evaluations of novel targeted therapies.

Recent advances in biotechnology and cancer genomics have afforded enormous opportunities for development of more effective anticancer therapies. A key thrust of this modern drug development paradigm is successful identification of predictive biomarkers that can distinguish patients who might be sensitive to new targeted therapies. To respond to this challenge, a number of phase III cancer trial designs integrating biomarker-based objectives have been proposed and implemented in oncology drug development. In this article, we provide an updated review of commonly used biomarker-based randomized clinical trial designs, with a particular focus on design efficiency. When the efficacy of a new therapy may be limited to a biomarker-defined subgroup, the choice of an appropriate randomized clinical trial design should be guided by the strength of the biomarker’s credentials. If compelling evidence indicates that a targeted therapy is beneficial only in a particular biomarker-defined subgroup, an enrichment design should be used. If there is strong evidence that the treatment is likely to be more beneficial in the biomarker-positive patients but a meaningful benefit is also possible in the biomarker-negative patients, then a properly powered biomarker-stratified design (eg, a subgroup-specific or Marker Sequential Test strategy) would provide the most rigorous determination of the sensitive populations. If the evidence supporting the predictive value of the biomarker is weak and the treatment is expected to work in the overall population, then a fallback design could be used. Careful selection of an appropriate phase III design strategy that integrates evaluation of a new anticancer therapy and its companion diagnostic is critical to the success of precision medicine in oncology.

Abstract Monoclonal gammopathies, including multiple myeloma (MM), represent a group of plasma cell (PC) disorders that comprise of mostly incurable hematopoietic malignancies with an increasing incidence in the US. Previous epidemiological studies demonstrated a 2-3 fold higher incidence of monoclonal gammopathy of undetermined significance (MGUS) and a similarly higher incidence of MM along with a ~4-year younger age of onset among African Americans (AA) compared to European Americans (EAs) (Fonseca, Leukemia, 2017). This suggests a possible ancestral-associated genetic predisposition of AAs to the development of monoclonal gammopathies (Landgren, Blood, 2006). When access to care is equal, AAs have a better overall survival compared to EAs suggesting that AAs may have a genetic predisposition that renders them better responders to treatment or have more favorable cytogenetic subtypes of MM (Waxman, Blood, 2010). Previous efforts to understand this disparity have relied on self-reported race rather than genetic ancestry. We hypothesize that quantifying genetic ancestry is necessary to fully understand the genetic mechanisms of racial disparities of monoclonal gammopathies. We describe our study of 881 patients with monoclonal gammopathies who had undergone uniform testing to identify MM-specific cytogenetic abnormalities. DNA from the bone marrow samples was genotyped on the Precision Medicine Research Array. Biogeographical ancestry was assessed using the Geographic Population Structure Origins tool, which provides the ancestral breakdown of 36 admixture components for each individual representing different geographic regions and outputs admixture proportions corresponding to those ancestries (Elhaik, Nat Commun, 2014). We evaluated whether an increase in the percentage of African Ancestry altered the odds of any primary cytogenetic abnormality. We demonstrated that a 10% increase in the percentage of African ancestry was associated with a 6% increase in the odds of detecting either an t(11;14), t(14;16) or t(14;20) (odds ratio = 1.06, 95% CI: 1.02 - 1.11; p-value = 0.05). Although we observed an increase in the odds of each of the individual translocations with 10% increase in African ancestry, l t(11;14) (OR=1.03) t(14;16) (OR=1.11) and t(14;20) (OR=1.10), these three abnormalities were combined to achieve greater statistical power. The differences in cytogenetic abnormalities were most striking among individuals with ≥80.0% African (n=120) compared to individuals with <0.1% African ancestry (n=235). Using these cutoff values comparing the ≥80.0% and <0.1% African ancestry cohorts, there was a higher prevalence of t(11;14), t(14;16) and t(14;20) (51% vs. 33%, respectively, p-value=0.008) and a lower prevalence of trisomies (with or without IgH translocations) (48.3% vs. 61.3%, respectively, p-value=0.066). In addition, the ≥80% African ancestry cohort also displayed a slightly lower prevalence of monosomy 13/13q deletion compared to the cohort with <0.1% African ancestry (34.2% vs. 38.7%, respectively, p-value=0.021). Further, we observed a significantly higher prevalence in the proportion of females with monoclonal gammopathies among the ≥80% African ancestry cohort compared to the cohort with <0.1% African ancestry (56.7% vs. 42.1%, respectively, p-value=0.028). We complement our past studies that relied on self-reported race and characterized patients' demographic and uniformly collected cytogenetic data in relation to genetically defined African ancestry. We demonstrate the prevalence of having one of three specific subtypes (i.e., t(11;14), t(14;16), or t(14;20)) was significantly higher in those with highest (≥80%) compared to lowest (<0.1%) African ancestry demonstrating African ancestry is associated with a subtype distribution of MM that differs from non-African individuals. These differences were only revealed after analysis of individuals with the highest and lowest percentage of African ancestry. Since 38.3% of the individuals with ≥80% African ancestry have the favorable t(11;14) translocation (in comparison to 26.8% of <0.1% African individuals), this finding may explain improved survival with equal access to care. Understanding the genetic mechanisms of this disparity is a fundamental step to understanding differences in incidence and outcomes of patients with monoclonal gammopathies. Disclosures Elhaik: DNA Diagnostics Center: Consultancy. Baird:DNA Diagnostics Center: Employment. Cerhan:Jannsen: Other: Scientific Advisory Board; Nanostring: Research Funding; Celgene: Research Funding. Stewart:Amgen Inc., BMS, Celgene, Takeda, Roche, Seattle Genetics, Janssen, Ono: Consultancy; Amgen Inc., Celgene, Roche, Seattle Genetics: Research Funding. Dispenzieri:Celgene, Takeda, Prothena, Jannsen, Pfizer, Alnylam, GSK: Research Funding. Kumar:Novartis: Research Funding; KITE: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; Takeda: Membership on an entity's Board of Directors or advisory committees; AbbVie: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; KITE: Membership on an entity's Board of Directors or advisory committees, Research Funding; Merck: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Oncopeptides: Membership on an entity's Board of Directors or advisory committees; AbbVie: Membership on an entity's Board of Directors or advisory committees, Research Funding; Roche: Research Funding.

PURPOSE To compare the prevalence of germline mutations in metastatic hormone-sensitive prostate cancer (mHSPC) and metastatic castrate-resistant prostate cancer (mCRPC) and assess the impact of mutations on progression to castration resistance and overall survival. METHODS Targeted sequencing of germline DNA from 704 men (221 at the time of mHSPC and 483 at the time of mCRPC) enrolled in two advanced prostate cancer registries at Mayo Clinic between 2003 and 2013 was performed for 21 predisposition genes. Frequencies of pathogenic mutations were compared in patients and reference controls to identify genes enriched in metastatic prostate cancer. Multivariable Cox proportional hazards regression was used to identify predictors of progression to mCRPC and overall survival. RESULTS Sixty-eight germline mutations in 12 genes were identified in 66 men (9.4%). Mutations in ATM, BRCA2, CHEK2, FANCM, and TP53 were significantly enriched (odds ratio greater than 2.0) in the metastatic cohorts compared with reference controls. The frequency of germline mutations was similar for patients with mHSPC and mCRPC (11.8% v 8.3%; P = .16). The median time to progression from mHSPC to mCRPC was 23.1 and 32.5 months for patients with and without mutations, respectively ( P = .96). Although older age at diagnosis, Gleason score greater than 7, elevated alkaline phosphatase level, and high volume of disease were associated with shorter duration of progression to mCRPC and poor overall survival, mutation status was not (progression to mCRPC hazard ratio, 0.81; 95% CI, 0.61 to 1.09; P = .17; overall survival hazard ratio, 1.00; 95% CI, 0.75 to 1.34; P = .98). CONCLUSION Similarly elevated rates of germline predisposition gene mutations in mHSPC and mCRPC suggest that germline genetic testing may help to guide medical management for all patients with advanced metastatic prostate cancer. Mutation status was not associated with shorter progression to mCRPC or poor overall survival.