Artificial Intelligence in Oncology: Current Landscape, Challenges, and Future Directions.

Artificial intelligence (AI) in oncology is advancing beyond algorithm development to integration into clinical practice. This review describes the current state of the field, with a specific focus on clinical integration. AI applications are structured according to cancer type and clinical domain, focusing on the four most common cancers and tasks of detection, diagnosis, and treatment. These applications encompass various data modalities, including imaging, genomics, and medical records. We conclude with a summary of existing challenges, evolving solutions, and potential future directions for the field. SIGNIFICANCE: AI is increasingly being applied to all aspects of oncology, where several applications are maturing beyond research and development to direct clinical integration. This review summarizes the current state of the field through the lens of clinical translation along the clinical care continuum. Emerging areas are also highlighted, along with common challenges, evolving solutions, and potential future directions for the field.

Incidence and Presenting Characteristics of Angiosarcoma in the US, 2001-2020.

Importance: Angiosarcoma is an aggressive vascular malignant neoplasm presenting either as a primary or secondary cancer, often arising after radiotherapy or in the context of preexisting lymphedema. Comprehensive data describing its incidence and presentation patterns are needed. Objective: To describe the incidence, presenting characteristics, and change over time of angiosarcoma in the US. Design, Setting, and Participants: This retrospective cross-sectional study used data from the US Cancer Statistics (USCS) National Program of Cancer Registries-Surveillance, Epidemiology, and End Results Combined Database, which captures more than 99% of newly diagnosed cancers in the US. The study included all 19 289 patients in the US with a new diagnosis of angiosarcoma between 2001 and 2020 captured in the USCS database. Statistical analysis was performed from June to September 2023. Main Outcomes and Measures: Incidence of angiosarcoma, demographics of patients with angiosarcoma, and extent of disease at presentation. Results: The study included 19 289 patients (median age, 71 years [IQR, 59-80 years]; 10 506 women [54.5%]) with a new diagnosis of angiosarcoma. The US incidence of angiosarcoma doubled between 2001 (657 cases) and 2019 (1312 cases), reflecting both an increase in the adjusted incidence rate of 1.6% per year (P = .001), to 3.3 cases per 1 000 000 person-years (95% CI, 3.1-3.5 cases per 1 000 000 person-years), and an increase in the population at risk. In 2020, the reported incidence rate (3.0 cases per 1 000 000 person-years) and cases of angiosarcoma (n = 1159) were modestly lower than in 2019. Overall, 72.3% of cases of angiosarcoma (n = 13 955) were cutaneous, subcutaneous, or breast angiosarcomas; 24.4% were visceral (n = 4701); and 3.3% were located in unknown or rare primary sites (n = 633). Secondary breast and chest wall angiosarcomas among women represented the largest contribution to increasing incidence. Among breast angiosarcomas, 99.2% (2684 of 2705) were in women and 71.9% (1944 of 2705) were secondary. A total of 80.4% of chest wall or thorax cases among women (1861 of 2316) were secondary vs 26.5% among men (112 of 422), and 63.9% of upper extremity cases among women (205 of 321) were secondary vs 26.8% (56 of 209) among men (P = .001). Rates of secondary angiosarcoma in the abdomen and lower extremities were similar between men and women. The incidence rate of visceral angiosarcoma was also found to be increasing (1.5% per year; P = .001). Conclusions and Relevance: This cross-sectional study describes angiosarcoma presentation patterns and incidence rates in the US over a 20-year period and shows that the number of cases in men and women increased, with the greatest increase among women with secondary angiosarcoma of the chest, breast, and upper extremity. These data increase awareness of a rare but highly morbid disease and highlight the need for improved early detection of angiosarcoma among patients at high risk, such as women with a history of breast cancer.

Effects of a change in recall period on reporting severe symptoms: an analysis of a pragmatic multisite trial.

BACKGROUND: Optimal methods for deploying electronic patient-reported outcomes to manage symptoms in routine oncologic practice remain uncertain. The electronic symptom management (eSyM) program asks chemotherapy and surgery patients to self-report 12 common symptoms regularly. Feedback from nurses and patients led to changing the recall period from the past 7 days to the past 24 hours. METHODS: Using questionnaires submitted during the 16 weeks surrounding the recall period change, we assessed the likelihood of reporting severe or moderate and severe symptoms across 12 common symptoms and separately for the 5 most prevalent symptoms. Interrupted time-series analyses modeled the effects of the change using generalized linear mixed-effects models. Surgery and chemotherapy cohorts were analyzed separately. Study-wide effects were estimated using a meta-analysis method. RESULTS: In total, 1692 patients from 6 institutions submitted 7823 eSyM assessments during the 16 weeks surrounding the recall period change. Shortening the recall period was associated with lower odds of severe symptom reporting in the surgery cohort (odds ratio = 0.65, 95% confidence interval = 0.46 to 0.93; P = .02) and lower odds of moderate and severe symptom reporting in the chemotherapy cohort (odds ratio = 0.83, 95% confidence interval = 0.71 to 0.97; P = .02). Among the most prevalent symptoms, 24-hour recall was associated with a lower rate of reporting postoperative constipation but no differences in reporting rates for other symptoms. CONCLUSION: A shorter recall period was associated with a reduction in the proportion of patients reporting moderate-severe symptoms. The optimal recall period may vary depending on whether electronic patient-reported outcomes are collected for active symptom management, as a clinical trial endpoint, or another purpose. ClinicalTrials.gov ID NCT03850912.

Identifying Oncology Clinical Trial Candidates Using Artificial Intelligence Predictions of Treatment Change: A Pilot Implementation Study.

PURPOSE: Precision oncology clinical trials often struggle to accrue, partly because it is difficult to find potentially eligible patients at moments when they need new treatment. We piloted deployment of artificial intelligence tools to identify such patients at a large academic cancer center. PATIENTS AND METHODS: Neural networks that process radiology reports to identify patients likely to start new systemic therapy were applied prospectively for patients with solid tumors that had undergone next-generation sequencing at our center. Model output was linked to the MatchMiner tool, which matches patients to trials using tumor genomics. Reports listing genomically matched patients, sorted by probability of treatment change, were provided weekly to an oncology nurse navigator (ONN) coordinating recruitment to nine early-phase trials. The ONN contacted treating oncologists when patients likely to change treatment appeared potentially trial-eligible. RESULTS: Within weekly reports to the ONN, 60,199 patient-trial matches were generated for 2,150 patients on the basis of genomics alone. Of these, 3,168 patient-trial matches (5%) corresponding to 525 patients were flagged for ONN review by our model, representing a 95% reduction in review compared with manual review of all patient-trial matches weekly. After ONN review for potential eligibility, treating oncologists for 74 patients were contacted. Common reasons for not contacting treating oncologists included cases where patients had already decided to continue current treatment (21%); the trial had no slots (14%); or the patient was ineligible on ONN review (12%). Of 74 patients whose oncologists were contacted, 10 (14%) had a consult regarding a trial and five (7%) enrolled. CONCLUSION: This approach facilitated identification of potential patients for clinical trials in real time, but further work to improve accrual must address the many other barriers to trial enrollment in precision oncology research.

Mortality Risks Over 20 Years in Men With Stage I to III Hormone Receptor-Positive Breast Cancer.

Importance: In women with hormone receptor-positive (HR+) breast cancer, the risk of distant recurrence and death persists for at least 20 years from diagnosis. The risk of late mortality in men with HR+ breast cancer has not been reported. Objective: To report 20-year risks of breast cancer-specific mortality (BCSM) and non-BCSM in men with stage I to III HR+ breast cancer and identify factors associated with late BCSM. Design, Setting, and Participants: An observational cohort study was conducted of men diagnosed with HR+ breast cancer from 1990 to 2008, using population-based data from the Surveillance, Epidemiology, and End Results program. Men diagnosed with stage I to III HR+ breast cancer were included in the analysis. Cumulative incidence function was used to estimate the outcomes of baseline clinicopathologic variables regarding cumulative risk of BCSM and non-BCSM since diagnosis. Smoothed hazard estimates over time were plotted for BCSM. Fine and Gray multivariable regression evaluated the association of preselected variables with BCSM, conditional on having survived 5 years. Main Outcome Measure: BCSM. Results: A total of 2836 men with stage I to III HR+ breast cancer were included, with a median follow-up of 15.41 (IQR, 12.08-18.67) years. Median age at diagnosis was 67 (IQR, 57-76) years. The cumulative 20-year risk of BCSM was 12.4% for stage I, 26.2% for stage II, and 46.0% for stage III. Smoothed annual hazard estimates for BCSM revealed an increase in late hazard rates with each incremental node category, reaching a bimodal distribution in N3 and stage III, with each having peaks in hazard rates at 4 and 11 years. Among patients who survived 5 years from diagnosis, the adjusted BCSM risk was higher for those younger than 50 years vs older than 64 years, those with grade II or III/IV vs grade I tumors, and stage II or III vs stage I disease. Conclusions and Relevance: The findings of this study suggest that, in men with stage I to III HR+ breast cancer, the risk of BCSM persists for at least 20 years and depends on traditional clinicopathologic factors, such as age, tumor stage, and tumor grade. Among men with higher stages of disease, the kinetics of the BCSM risk appear different from the risk that has been reported in women.

Strategies for Implementing an Electronic Patient-Reported Outcomes-Based Symptom Management Program Across Six Cancer Centers.

Background: Electronic patient-reported outcome (ePRO)-based symptom management improves cancer patients' outcomes. However, implementation of ePROs is challenging, requiring technical resources for integration into clinical systems, substantial buy-in from clinicians and patients, novel workflows to support between-visit symptom management, and institutional investment. Methods: The SIMPRO Research Consortium developed eSyM, an electronic health record-integrated, ePRO-based symptom management program for medical oncology and surgery patients and deployed it at six cancer centers between August 2019 and April 2022 in a type II hybrid effectiveness-implementation cluster randomized stepped-wedge study. Sites documented implementation strategies monthly using REDCap, itemized them using the Expert Recommendations for Implementation Change (ERIC) list and mapped their target barriers using the Consolidated Framework for Implementation Research (CFIR) to inform eSyM program enhancement, facilitate inter-consortium knowledge sharing and guide future deployment efforts. Results: We documented 226 implementation strategies: 35 'foundational' strategies were applied consortium-wide by the coordinating center and 191 other strategies were developed by individual sites. We consolidated these 191 site-developed strategies into 64 unique strategies (i.e., removed duplicates) and classified the remainder as either 'universal', consistently used by multiple sites (N=29), or 'adaptive', used only by individual sites (N=35). Universal strategies were perceived as having the highest impact; they addressed eSyM clinical preparation, training, engagement of patients/clinicians, and program evaluation. Across all documented SIMPRO strategies, 44 of the 73 ERIC strategies were addressed and all 5 CFIR barriers were addressed. Conclusion: Methodical collection of theory-based implementation strategies fostered the identification of universal, high-impact strategies that facilitated adoption of a novel care-delivery intervention by patients, clinicians, and institutions. Attention to the high-impact strategies identified in this project could support implementation of ePROs as a component of routine cancer care at other institutions.

Disparities in electronic health record portal access and use among patients with cancer.

BACKGROUND: Electronic health record-linked portals may improve health-care quality for patients with cancer. Barriers to portal access and use undermine interventions that rely on portals to reduce cancer care disparities. This study examined portal access and persistence of portal use and associations with patient and structural factors before the implementation of 3 portal-based interventions within the Improving the Management of symPtoms during And following Cancer Treatment (IMPACT) Consortium. METHODS: Portal use data were extracted from electronic health records for the 12 months preceding intervention implementation. Sociodemographic factors, mode of accessing portals (web vs mobile), and number of clinical encounters before intervention implementation were also extracted. Rurality was derived using rural-urban commuting area codes. Broadband access was estimated using the 2015-2019 American Community Survey. Multiple logistic regression models tested the associations of these factors with portal access (ever accessed or never accessed) and persistence of portal use (accessed the portal ≤20 weeks vs ≥21 weeks in the 35-week study period). RESULTS: Of 28 942 eligible patients, 10 061 (35%) never accessed the portal. Male sex, membership in a racial and ethnic minority group, rural dwelling, not working, and limited broadband access were associated with lower odds of portal access. Younger age and more clinical encounters were associated with higher odds of portal access. Of those with portal access, 25% were persistent users. Using multiple modalities for portal access, being middle-aged, and having more clinical encounters were associated with persistent portal use. CONCLUSION: Patient and structural factors affect portal access and use and may exacerbate disparities in electronic health record-based cancer symptom surveillance and management.

Assessing Patient Readiness for an Electronic Patient-Reported Outcome-Based Symptom Management Intervention in a Multisite Study.

PURPOSE: While the use of electronic patient-reported outcomes (ePROs) in routine clinical practice is increasing, barriers to patient engagement limit adoption. Studies have focused on technology access as a key barrier, yet other characteristics may also confound readiness to use ePROs including patients' confidence in using technology and confidence in asking clinicians questions. METHODS: To assess readiness to use ePROs, adult patients from six US-based health systems who started a new oncology treatment or underwent a cancer-directed surgery were invited to complete a survey that assessed access to and confidence in the use of technology, ease of asking clinicians questions about health, and symptom management self-efficacy. Multivariable ordinal logistic regression models were fit to assess the association between technology confidence, ease of asking questions, and symptom management self-efficacy. RESULTS: We contacted 3,212 individuals, and 1,043 (33%) responded. The median age was 63 years, 68% were female, and 75% reported having access to patient portals. Over 80% had two or more electronic devices. Most patients reported high technology confidence, higher ease of asking clinicians questions, and high symptom management self-efficacy (n = 692; 66%). Patients with high technology confidence also reported higher ease of asking nurses about their health (adjusted odds ratio [AOR], 4.58 [95% CI, 2.36 to 8.87]; P ≤ .001). Those who reported higher ease of asking nurses questions were more likely to report higher confidence in managing symptoms (AOR, 30.54 [95% CI, 12.91 to 72.30]; P ≤ .001). CONCLUSION: Patient readiness to use ePROs likely depends on multiple factors, including technology and communication confidence, and symptom management self-efficacy. Future studies should assess interventions to address these factors.

The Longitudinal Implementation Strategy Tracking System (LISTS): feasibility, usability, and pilot testing of a novel method.

BACKGROUND: Systematic approaches are needed to accurately characterize the dynamic use of implementation strategies and how they change over time. We describe the development and preliminary evaluation of the Longitudinal Implementation Strategy Tracking System (LISTS), a novel methodology to document and characterize implementation strategies use over time. METHODS: The development and initial evaluation of the LISTS method was conducted within the Improving the Management of SymPtoms during And following Cancer Treatment (IMPACT) Research Consortium (supported by funding provided through the NCI Cancer MoonshotSM). The IMPACT Consortium includes a coordinating center and three hybrid effectiveness-implementation studies testing routine symptom surveillance and integration of symptom management interventions in ambulatory oncology care settings. LISTS was created to increase the precision and reliability of dynamic changes in implementation strategy use over time. It includes three components: (1) a strategy assessment, (2) a data capture platform, and (3) a User's Guide. An iterative process between implementation researchers and practitioners was used to develop, pilot test, and refine the LISTS method prior to evaluating its use in three stepped-wedge trials within the IMPACT Consortium. The LISTS method was used with research and practice teams for approximately 12 months and subsequently we evaluated its feasibility, acceptability, and usability using established instruments and novel questions developed specifically for this study. RESULTS: Initial evaluation of LISTS indicates that it is a feasible and acceptable method, with content validity, for characterizing and tracking the use of implementation strategies over time. Users of LISTS highlighted several opportunities for improving the method for use in future and more diverse implementation studies. CONCLUSIONS: The LISTS method was developed collaboratively between researchers and practitioners to fill a research gap in systematically tracking implementation strategy use and modifications in research studies and other implementation efforts. Preliminary feedback from LISTS users indicate it is feasible and usable. Potential future developments include additional features, fewer data elements, and interoperability with alternative data entry platforms. LISTS offers a systematic method that encourages the use of common data elements to support data analysis across sites and synthesis across studies. Future research is needed to further adapt, refine, and evaluate the LISTS method in studies with employ diverse study designs and address varying delivery settings, health conditions, and intervention types.

Systematic symptom management in the IMPACT Consortium: rationale and design for 3 effectiveness-implementation trials.

Cancer and its treatment produce deleterious symptoms across the phases of care. Poorly controlled symptoms negatively affect quality of life and result in increased health-care needs and hospitalization. The Improving the Management of symPtoms during And following Cancer Treatment (IMPACT) Consortium was created to develop 3 large-scale, systematic symptom management systems, deployed through electronic health record platforms, and to test them in pragmatic, randomized, hybrid effectiveness and implementation trials. Here, we describe the IMPACT Consortium's conceptual framework, its organizational components, and plans for evaluation. The study designs and lessons learned are highlighted in the context of disruptions related to the COVID-19 pandemic.

Feasibility of Weekly Electronic Health Record-Embedded Patient-Reported Outcomes for Patients Starting Oral Cancer-Directed Therapy.

PURPOSE: To examine the feasibility of integrating a symptom management platform into the electronic health record (EHR) using electronic patient-reported outcomes (ePROs) during oral cancer-directed therapy (OCDT) and explore the impact of prompting oncology nurse navigators (ONNs) to respond to severe symptomatic adverse events (SAEs). MATERIALS AND METHODS: Adults prescribed OCDT at Dana-Farber Cancer Institute were consecutively invited to participate. Participants received weekly messages to complete ePROs. The first half enrolled in a passive (P) group where ePROs responses could be viewed anytime, but outreach was not expected. The second half enrolled in an active (A) group where severe SAEs prompted emails to ONNs for outreach within 1 business day. Feasibility was the proportion of participants completing ≥2 ePROs during the first 30 days. Participants were followed for up to 90 days. RESULTS: From June 25, 2019, to August 18, 2021, 100 participants enrolled, and 96 remained enrolled for at least 30 days. Overall, average age was 59 years, 80% female, and 9% used the platform in Spanish. Twenty-two A (45%) and 27 P (57%) participants met the feasibility threshold (P = .26). ePROs returned at 30 days were similar (P = .50): 0 ePROs 17 A, 13 P; 1 ePRO 10 A, 7 P; 2 ePROs 3 A, 5 P; 3 ePROs 1 A, 4 P; 4 ePROs 7 A, 8 P; and 5 ePROs 11 A, 10 P. Documented telephone encounters at 30 days were similar (109 A, 101 P; P = .86). CONCLUSION: EHR-embedded ePROs administered weekly for people on OCDT was feasible, although many went incomplete. ePRO completion was not clearly affected by nursing calls for severe SAEs. Future efforts will investigate improving engagement and addressing symptoms proactively.

Machine learning for genetics-based classification and treatment response prediction in cancer of unknown primary.

Cancer of unknown primary (CUP) is a type of cancer that cannot be traced back to its primary site and accounts for 3-5% of all cancers. Established targeted therapies are lacking for CUP, leading to generally poor outcomes. We developed OncoNPC, a machine-learning classifier trained on targeted next-generation sequencing (NGS) data from 36,445 tumors across 22 cancer types from three institutions. Oncology NGS-based primary cancer-type classifier (OncoNPC) achieved a weighted F1 score of 0.942 for high confidence predictions ([Formula: see text]) on held-out tumor samples, which made up 65.2% of all the held-out samples. When applied to 971 CUP tumors collected at the Dana-Farber Cancer Institute, OncoNPC predicted primary cancer types with high confidence in 41.2% of the tumors. OncoNPC also identified CUP subgroups with significantly higher polygenic germline risk for the predicted cancer types and with significantly different survival outcomes. Notably, patients with CUP who received first palliative intent treatments concordant with their OncoNPC-predicted cancers had significantly better outcomes (hazard ratio (HR) = 0.348; 95% confidence interval (CI) = 0.210-0.570; P = [Formula: see text]). Furthermore, OncoNPC enabled a 2.2-fold increase in patients with CUP who could have received genomically guided therapies. OncoNPC thus provides evidence of distinct CUP subgroups and offers the potential for clinical decision support for managing patients with CUP.

Estimating mortality in women with triple-negative breast cancer: The 'ESTIMATE triple-negative' tool.

PURPOSE: Triple-negative breast cancer (TNBC) is associated with a high risk of breast cancer-specific mortality (BCSM). Estimating the risk of BCSM and non-BCSM in TNBC would aid clinical decision-making. We developed the tool 'ESTIMATE-TN', to assess BCSM, non-BCSM, and all-cause mortality in non-metastatic TNBC. METHODS: Using Surveillance, Epidemiology, and End Results (SEER), we created an interactive tool that provides a nonparametric estimate of the cumulative risk of BCSM and non-BCSM between years 0 and 7 from diagnosis, accounting for baseline clinical and pathologic variables, using Gray's subdistribution method. RESULTS: We included 37,293 women with TNBC diagnosed during 2010-2017. Most patients were White (71.9%) and aged 50-69 years (51.3%). Most tumour characteristics were high-grade (78.6%), T2 (42.4%), and N0 (69.5%). ESTIMATE-TN allows to input patient and tumour characteristics, and the preferred timeframe. For example, patients aged 50-59 years with a new diagnosis of T2, N1, high-grade TNBC have a risk of BCSM at 7 years of 30.8% (95% confidence interval [CI]: 26.3-35.4%) and a risk of non-BCSM over the same period of 2.8% (95% CI: 1.3-4.3%). After 3 years from initial diagnosis, the residual cumulative risks of BCSM and non-BCSM at 7 years are 17.4% (95% CI: 12.6-22.2%) and 1.1% (95% CI: 0-2.5%), respectively. CONCLUSIONS: ESTIMATE-TN is an interactive tool for TNBC that can be used to integrate population-based risks of BCSM and non-BCSM based on patient and tumour characteristics, facilitating our understanding of competing risks of death, which can aid clinical decision-making.

Association of Variation in US County-Level Rates of Liver Surgical Resection for Colorectal Liver Metastasis With Poverty Rates in 2010.

Importance: Among patients with colorectal liver metastasis (CRLM) who are eligible for curative-intent liver surgical resection, only half undergo liver metastasectomy. It is currently unclear how rates of liver metastasectomy vary geographically in the US. Geographic differences in county-level socioeconomic characteristics may, in part, explain variability in the receipt of liver metastasectomy for CRLM. Objective: To describe county-level variation in the receipt of liver metastasectomy for CRLM in the US and its association with poverty rates. Design, Setting, and Participants: This ecological, cross-sectional, and county-level analysis was conducted using data from the Surveillance, Epidemiology, and End Results Research Plus database. The study included the county-level proportion of patients who had colorectal adenocarcinoma diagnosed between January 1, 2010, and December 31, 2018, underwent primary surgical resection, and had liver metastasis without extrahepatic metastasis. The county-level proportion of patients with stage I colorectal cancer (CRC) was used as a comparator. Data analysis was performed on March 2, 2022. Exposures: County-level poverty in 2010 obtained from the US Census (proportion of county population below the federal poverty level). Main Outcomes and Measures: The primary outcome was county-level odds of liver metastasectomy for CRLM. The comparator outcome was county-level odds of surgical resection for stage I CRC. Multivariable binomial logistic regression accounting for clustering of outcomes within a county via an overdispersion parameter was used to estimate the county-level odds of receiving a liver metastasectomy for CRLM associated with a 10% increase in poverty rate. Results: In the 194 US counties included in this study, there were 11 348 patients. At the county level, the majority of the population was male (mean [SD], 56.9% [10.2%]), White (71.9% [20.0%]), and aged between 50 and 64 (38.1% [11.0%]) or 65 and 79 (33.6% [11.4%]) years. The adjusted odds of undergoing a liver metastasectomy was lower in counties with higher poverty in 2010 (per 10% increase; odds ratio, 0.82 [95% CI, 0.69-0.96]; P = .02). County-level poverty was not associated with receipt of surgery for stage I CRC. Despite the difference in rates of surgery (mean county-level rates were 0.24 for liver metastasectomy for CRLM and 0.75 for surgery for stage I CRC), the variance at the county-level for these 2 surgical procedures was similar (F370, 193 = 0.81; P = .08). Conclusions and Relevance: The findings of this study suggest that higher poverty was associated with lower receipt of liver metastasectomy among US patients with CRLM. Surgery for a more common and less complex cancer comparator (ie, stage I CRC) was not observed to be associated with county-level poverty rates. However, county-level variation in surgical rates was similar for CRLM and stage I CRC. These findings further suggest that access to surgical care for complex gastrointestinal cancers such as CRLM may be partially influenced by where patients live.

Utilizing Electronic Health Records (EHR) and Tumor Panel Sequencing to Demystify Prognosis of Cancer of Unknown Primary (CUP) patients.

Cancer of unknown primary (CUP) is a type of cancer that cannot be traced back to its original site and accounts for 3-5% of all cancers. It does not have established targeted therapies, leading to poor outcomes. We developed OncoNPC, a machine learning classifier trained on targeted next-generation sequencing data from 34,567 tumors from three institutions. OncoNPC achieved a weighted F1 score of 0.94 for high confidence predictions on known cancer types (65% of held-out samples). When applied to 971 CUP tumors from patients treated at the Dana-Farber Cancer Institute, OncoNPC identified actionable molecular alterations in 23% of the tumors. Furthermore, OncoNPC identified CUP subtypes with significantly higher polygenic germline risk for the predicted cancer type and significantly different survival outcomes, supporting its validity. Importantly, CUP patients who received first palliative intent treatments concordant with their OncoNPC-predicted cancer sites had significantly better outcomes (H.R. 0.348, 95% C.I. 0.210 - 0.570, p-value 2.32 × 10-5). OncoNPC thus provides evidence of distinct CUP subtypes and offers the potential for clinical decision support for managing patients with CUP.

Programmatic Precision Oncology Decision Support for Patients With Gastrointestinal Cancer.

PURPOSE: With the growing number of available targeted therapeutics and molecular biomarkers, the optimal care of patients with cancer now depends on a comprehensive understanding of the rapidly evolving landscape of precision oncology, which can be challenging for oncologists to navigate alone. METHODS: We developed and implemented a precision oncology decision support system, GI TARGET, (Gastrointestinal Treatment Assistance Regarding Genomic Evaluation of Tumors) within the Gastrointestinal Cancer Center at the Dana-Farber Cancer Institute. With a multidisciplinary team, we systematically reviewed tumor molecular profiling for GI tumors and provided molecularly informed clinical recommendations, which included identifying appropriate clinical trials aided by the computational matching platform MatchMiner, suggesting targeted therapy options on or off the US Food and Drug Administration-approved label, and consideration of additional or orthogonal molecular testing. RESULTS: We reviewed genomic data and provided clinical recommendations for 506 patients with GI cancer who underwent tumor molecular profiling between January and June 2019 and determined follow-up using the electronic health record. Summary reports were provided to 19 medical oncologists for patients with colorectal (n = 198, 39%), pancreatic (n = 124, 24%), esophagogastric (n = 67, 13%), biliary (n = 40, 8%), and other GI cancers. We recommended ≥ 1 precision medicine clinical trial for 80% (406 of 506) of patients, leading to 24 enrollments. We recommended on-label and off-label targeted therapies for 6% (28 of 506) and 25% (125 of 506) of patients, respectively. Recommendations for additional or orthogonal testing were made for 42% (211 of 506) of patients. CONCLUSION: The integration of precision medicine in routine cancer care through a dedicated multidisciplinary molecular tumor board is scalable and sustainable, and implementation of precision oncology recommendations has clinical utility for patients with cancer.

Clinicopathologic and sociodemographic factors associated with late relapse triple negative breast cancer in a multivariable logistic model: A multi-institution cohort study.

BACKGROUND: Most metastatic recurrences of triple negative breast cancer (TNBC) occur within five years of diagnosis, yet late relapses of TNBC (lrTNBC) do occur. Our objective was to develop a risk prediction model of lrTNBC using readily available clinicopathologic and sociodemographic features. METHODS: We included patients diagnosed with stage I-III TNBC between 1998 and 2012 at ten academic cancer centers. lrTNBC was defined as relapse or mortality greater than 5 years from diagnosis. Features associated with lrTNBC were included in a multivariable logistic model using backward elimination with a p < 0.10 criterion, with a final multivariable model applied to training (70%) and independent validation (30%) cohorts. RESULTS: A total 2210 TNBC patients with at least five years follow-up and no relapse before 5 years were included. In final multivariable model, lrTNBC was significantly associated with higher stage at diagnosis (adjusted Odds Ratio [aOR] for stage III vs I, 10.9; 95% Confidence Interval [CI], 7.5-15.9; p < 0.0001) and BMI (aOR for obese vs normal weight, 1.4; 95% CI, 1.0-1.8; p = 0.03). Final model performance was consistent between training (70%) and validation (30%) cohorts. CONCLUSIONS: A risk prediction model incorporating stage, BMI, and age at diagnosis offers potential utility for identification of patients at risk of development of lrTNBC and warrants further investigation.

Survival in male breast cancer over the past 3 decades.

BACKGROUND: Breast cancer mortality in women has declined statistically significantly over the past several years. In men, it is unclear whether survival has changed over time. We evaluated changes in breast cancer-specific survival (BCSS) and overall survival (OS) in male breast cancer over the past 3 decades. METHODS: We evaluated men diagnosed with breast cancer between 1988 and 2017, reported in the Surveillance, Epidemiology, and End Results registry. Patients were categorized into 3 groups by year of diagnosis: 1988-1997, 1998-2007, and 2008-2017. BCSS and OS were estimated by Kaplan-Meier, and differences between groups were compared by log-rank test. Multivariable Cox regression evaluated the independent association of year of diagnosis with BCSS and OS. All tests were 2-sided. RESULTS: We included 8481 men. Overall, BCSS at 5 years was 83.69%, 83.78%, and 84.41% in groups 1988-1997, 1998-2007, and 2008-2017, respectively (P = .86). There was no statistically significant difference in BCSS between the 3 groups within each stage of disease. Among all patients, OS at 5 years was 64.61%, 67.31%, and 69.05% in groups 1988-1997, 1998-2007, and 2008-2017, respectively (P = .01). In adjusted Cox models, each additional year of diagnosis had no statistically significant association with BCSS (hazard ratio = 1.00, 95% confidence interval = 0.99 to 1.01, P = .75), but there was statistically significant improvement in OS (hazard ratio = 0.99, 95% CI = 0.98 to 0.99, P = .009). CONCLUSIONS: Over the past 3 decades, there has been no statistically significant improvement in BCSS in male breast cancer. Changes in OS over time are consistent with increasing life expectancy. Efforts to improve BCSS in male breast cancer are warranted.

Regional Disparities in the Use and Delivery of Adjuvant Radiation Therapy after Lumpectomy for Breast Cancer in the Medicare Population.

Purpose: We examined radiation therapy (RT) use among patients with early-stage breast cancer and analyzed the contribution of patient, cancer, and regional factors to the likelihood of RT receipt across Health Service Areas. Methods and Materials: We identified 13,176 patients aged 66 to 79 years in the Surveillance, Epidemiology, and End Results (SEER) Program-Medicare database who were diagnosed with lymph node-negative breast cancer in 2007 to 2011 and were treated with breast-conserving surgery. Patients were stratified as being at high risk or low risk for recurrence based on National Comprehensive Cancer Network Guidelines. Receipt of RT was studied with 5 modeling approaches to determine whether RT use and regional variation in its use changed based on the risk level of the cohort. Multivariable mixed-effects logistic regression was performed for each outcome. Choropleth maps were used to describe patterns of RT use. Results: Among high-risk patients, 70.1% received RT, compared with 72.6% of low-risk patients (P = .002). Among patients receiving RT, 60.9% were classified as high-risk, compared with 63.0% of patients who did not receive RT (P = .002). In multivariable analyses, patients in all rural areas had lower odds of receiving RT compared with the entire cohort (odds ratio [OR], 0.73; P < .001) and had lower odds of being high-risk and receiving RT (OR, 0.69; P < .001). Black patients (OR, 0.73; P = .001) and Asian patients (OR, 0.74; P = .004) had decreased likelihood of receiving RT compared with the entire cohort. The regional interclass correlation coefficient (ICC) for the model predicting receipt of RT among all patients was 0.05 and among low-risk patients was 0.06. The regional ICC dropped to 0.02 for the model predicting being both high-risk and receiving RT among all patients. Conclusions: We observed regional and racial and ethnic disparities in RT receipt among our cohort. Reassuringly, less regional variability was observed for RT receipt among those at high risk for recurrence. Future work is needed to understand the causes of these regional disparities to better serve patients who may benefit from treatment.