Suicide Risk Screening for Head and Neck Cancer Patients: An Implementation Study.

OBJECTIVES: There is limited research on suicide risk screening (SRS) among head and neck cancer (HNC) patients, a population at increased risk for suicide. To address this gap, this single-site mixed methods study assessed oncology professionals' perspectives about the feasibility, acceptability, and appropriateness of an electronic SRS program that was implemented as a part of routine care for HNC patients. METHODS: Staff who assisted with SRS implementation completed (e.g., nurses, medical assistants, advanced practice providers, physicians, social workers) a one-time survey (N = 29) and interview (N = 25). Quantitative outcomes were assessed using previously validated feasibility, acceptability, and appropriateness measures. Additional qualitative data were collected to provide context for interpreting the scores. RESULTS: Nurses and medical assistants, who were directly responsible for implementing SRS, reported low feasibility, acceptability, and appropriateness, compared with other team members (e.g., physicians, social workers, advanced practice providers). Team members identified potential improvements needed to optimize SRS, such as hiring additional staff, improving staff training, providing different modalities for screening completion among individuals with disabilities, and revising the patient-reported outcomes to improve suicide risk prediction. CONCLUSION: Staff perspectives about implementing SRS as a part of routine cancer care for HNC patients varied widely. Before screening can be implemented on a larger scale for HNC and other cancer patients, additional implementation strategies may be needed that optimize workflow and reduce staff burden, such as staff training, multiple modalities for completion, and refined tools for identifying which patients are at greatest risk for suicide.

Melanoma and CLL co-occurrence and survival: role of KC history.

BACKGROUND: Survival following melanoma and chronic lymphocytic leukemia (CLL) have both been individually associated with previous history of non-melanoma skin cancers (specifically keratinocyte carcinomas [KC]). Furthermore, melanoma and CLL have been reported to occur within the same patients. The survival experience of patients with both cancers is understudied, and the role of history of KC is unknown. Additional research is needed to tease apart the independent associations between KC and CLL survival, KC and melanoma survival, and the co-occurrence of all three cancers. METHODS: A retrospective cohort study was conducted among patients who were diagnosed with melanoma and/or CLL at a comprehensive cancer center between 2008 and 2020. Multivariable Cox regression models were used to examine the association between history of KC and survival following melanoma and/or CLL with careful consideration of calendar year of diagnosis, treatment regimens and other risk factors. A nested case-control study comparing patients with both CLL and melanoma to those with only CLL or only melanoma was conducted to compare blood parameters across the three groups. RESULTS: A time-dependent association was observed between history of KC and favorable melanoma survival within 4 years following diagnosis and poorer survival post 7 years after melanoma diagnosis. History of KC was not significantly associated with survival following the diagnosis of CLL, after adjustment for clinical factors including historical/concurrent melanoma. Patients with co-occurring melanoma and CLL tended to be diagnosed with melanoma first and had elevated blood parameters including white blood cell and lymphocyte counts as compared with patients who were diagnosed with only melanoma. CONCLUSIONS: History of KC was an independent predictor of survival following melanoma but not of CLL. Additional studies are needed to determine if blood parameters obtained at the time of melanoma diagnosis could be used as a cost-effective way to identify those at high risk of asymptomatic CLL for the promotion of earlier CLL diagnosis.

Factors Associated with Self-reported COVID-19 Infection and Hospitalization among Patients Seeking Care at a Comprehensive Cancer Center.

BACKGROUND: COVID-19 infection severity differs by race and ethnicity, but its long-term effect on cancer-related outcomes is unknown. Therefore, information on COVID-19 history is critical to ascertain among new cancer patients in order to advance research on its impact on cancer outcomes and potentially related health disparities. METHODS: A cross-sectional study was conducted among 16,025 new patients seeking care at Moffitt Cancer Center (MCC) between 2021 and 2022. Patient self-reported histories of COVID-19 infection and other pre-existing health conditions were obtained from electronic questionnaires administered to all new MCC patients. Associations between demographics and COVID-19 infection and hospitalization were examined. RESULTS: A total of 1,971 patients (12.3%) reported ever having COVID-19. Self-reported COVID-19 history was significantly more prevalent in Hispanic vs. non-Hispanic patients (OR = 1.24, 1.05-1.45) and less prevalent in Asian versus White patients (OR = 0.49, 95% 0.33-0.70). Among patients who ever had COVID-19, 10.6% reported a COVID-19-related hospitalization. Males had higher odds of a COVID-19 related hospitalization than females (OR = 1.50, 95% CI = 1.09-2.05), as did Black/African American patients (OR = 2.11, 95% CI = 1.18-3.60) and patients of races other than Black/African American and Asian (OR = 2.61, 95% CI = 1.43-4.54) compared to White patients. Hispanic patients also experienced higher odds of hospitalization (OR = 2.06, 95% CI-1.29- 3.23) compared with non-Hispanic patients of all races in a sensitivity analysis that combined race/ethnicity. Pre-existing lung and breathing problems were associated with higher odds of being hospitalized with COVID-19 (OR = 2.38, 95% CI = 1.61-3.48), but these and other health conditions did not explain the observed associations between race and COVID-19 hospitalization. CONCLUSIONS: Higher rates of COVID-19 hospitalization were observed among patients identifying as Black/African American or Hispanic independent of pre-existing health conditions. Future studies evaluating long-term effects of COVID-19 should carefully examine potential racial/ethnic disparities in cancer outcomes.

Translation of AI into oncology clinical practice.

Artificial intelligence (AI) is a transformative technology that is capturing popular imagination and can revolutionize biomedicine. AI and machine learning (ML) algorithms have the potential to break through existing barriers in oncology research and practice such as automating workflow processes, personalizing care, and reducing healthcare disparities. Emerging applications of AI/ML in the literature include screening and early detection of cancer, disease diagnosis, response prediction, prognosis, and accelerated drug discovery. Despite this excitement, only few AI/ML models have been properly validated and fewer have become regulated products for routine clinical use. In this review, we highlight the main challenges impeding AI/ML clinical translation. We present different clinical use cases from the domains of radiology, radiation oncology, immunotherapy, and drug discovery in oncology. We dissect the unique challenges and opportunities associated with each of these cases. Finally, we summarize the general requirements for successful AI/ML implementation in the clinic, highlighting specific examples and points of emphasis including the importance of multidisciplinary collaboration of stakeholders, role of domain experts in AI augmentation, transparency of AI/ML models, and the establishment of a comprehensive quality assurance program to mitigate risks of training bias and data drifts, all culminating toward safer and more beneficial AI/ML applications in oncology labs and clinics.

Overview of approaches to estimate real-world disease progression in lung cancer.

BACKGROUND: Randomized clinical trials of novel treatments for solid tumors normally measure disease progression using the Response Evaluation Criteria in Solid Tumors. However, novel, scalable approaches to estimate disease progression using real-world data are needed to advance cancer outcomes research. The purpose of this narrative review is to summarize examples from the existing literature on approaches to estimate real-world disease progression and their relative strengths and limitations, using lung cancer as a case study. METHODS: A narrative literature review was conducted in PubMed to identify articles that used approaches to estimate real-world disease progression in lung cancer patients. Data abstracted included data source, approach used to estimate real-world progression, and comparison to a selected gold standard (if applicable). RESULTS: A total of 40 articles were identified from 2008 to 2022. Five approaches to estimate real-world disease progression were identified including manual abstraction of medical records, natural language processing of clinical notes and/or radiology reports, treatment-based algorithms, changes in tumor volume, and delta radiomics-based approaches. The accuracy of these progression approaches were assessed using different methods, including correlations between real-world endpoints and overall survival for manual abstraction (Spearman rank ρ = 0.61-0.84) and area under the curve for natural language processing approaches (area under the curve = 0.86-0.96). CONCLUSIONS: Real-world disease progression has been measured in several observational studies of lung cancer. However, comparing the accuracy of methods across studies is challenging, in part, because of the lack of a gold standard and the different methods used to evaluate accuracy. Concerted efforts are needed to define a gold standard and quality metrics for real-world data.

Discordant pathologic diagnoses of myelodysplastic neoplasms and their implications for registries and therapies.

Myelodysplastic neoplasms (MDS) are a collection of hematopoietic disorders with widely variable prognoses and treatment options. Accurate pathologic diagnoses present challenges because of interobserver variability in interpreting morphology and quantifying dysplasia. We compared local clinical site diagnoses with central, adjudicated review from 918 participants enrolled in the ongoing National Heart, Lung, and Blood Institute National MDS Natural History Study, a prospective observational cohort study of participants with suspected MDS or MDS/myeloproliferative neoplasms (MPNs). Locally, 264 (29%) were diagnosed as having MDS, 15 (2%) MDS/MPN overlap, 62 (7%) idiopathic cytopenia of undetermined significance (ICUS), 0 (0%) acute myeloid leukemia (AML) with <30% blasts, and 577 (63%) as other. Approximately one-third of cases were reclassified after central review, with 266 (29%) diagnosed as MDS, 45 (5%) MDS/MPN overlap, 49 (5%) ICUS, 15 (2%) AML with <30%, and 543 (59%) as other. Site miscoding errors accounted for more than half (53%) of the local misdiagnoses, leaving a true misdiagnosis rate of 15% overall, 21% for MDS. Therapies were reported in 37% of patients, including 43% of patients with MDS, 49% of patients with MDS/MPN, and 86% of patients with AML with <30% blasts. Treatment rates were lower (25%) in cases with true discordance in diagnosis compared with those for whom local and central diagnoses agreed (40%), and receipt of inappropriate therapy occurred in 7% of misdiagnosed cases. Discordant diagnoses were frequent, which has implications for the accuracy of study-related and national registries and can lead to inappropriate therapy. This trial was registered at www.clinicaltrials.gov as #NCT05074550.

Examining disparities in large-scale patient-reported data capture using digital tools among cancer patients at clinical intake.

BACKGROUND: Patient-reported data can improve quality of healthcare delivery and patient outcomes. Moffitt Cancer Center ("Moffitt") administers the Electronic Patient Questionnaire (EPQ) to collect data on demographics, including sexual orientation and gender identity (SOGI), medical history, cancer risk factors, and quality of life. Here we investigated differences in EPQ completion by demographic and cancer characteristics. METHODS: An analysis including 146,142 new adult patients at Moffitt in 2009-2020 was conducted using scheduling, EPQ and cancer registry data. EPQ completion was described by calendar year and demographics. Logistic regression was used to estimate associations between demographic/cancer characteristics and EPQ completion. More recently collected information on SOGI were described. RESULTS: Patient portal usage (81%) and EPQ completion rates (79%) were consistently high since 2014. Among patients in the cancer registry, females were more likely to complete the EPQ than males (odds ratio [OR] = 1.17, 95% confidence interval [CI] = 1.14-1.20). Patients ages 18-64 years were more likely to complete the EPQ than patients aged ≥65. Lower EPQ completion rates were observed among Black or African American patients (OR = 0.59, 95% CI = 0.56-0.63) as compared to Whites and among patients whose preferred language was Spanish (OR = 0.40, 95% CI = 0.36-0.44) or another language as compared to English. Furthermore, patients with localized (OR = 1.16, 95% CI = 1.12-1.19) or regional (OR = 1.16, 95% CI = 1.12-1.20) cancer were more likely to complete the EPQ compared to those with metastatic disease. Less than 3% of patients self-identified as being lesbian, gay, or bisexual and <0.1% self-identified as transgender, genderqueer, or other. CONCLUSIONS: EPQ completion rates differed across demographics highlighting opportunities for targeted process improvement. Healthcare organizations should evaluate data acquisition methods to identify potential disparities in data completeness that can impact quality of clinical care and generalizability of self-reported data.

Telemedicine Adoption in an NCI-Designated Cancer Center During the COVID-19 Pandemic: A Report on Patient Experience of Care.

BACKGROUND: Patients with cancer require timely access to care so that healthcare providers can prepare an optimal treatment plan with significant implications for quality of life and mortality. The COVID-19 pandemic spurred rapid adoption of telemedicine in oncology, but study of patient experience of care with telemedicine in this population has been limited. We assessed overall patient experience of care with telemedicine at an NCI-designated Comprehensive Cancer Center during the COVID-19 pandemic and examined changes in patient experience over time. PATIENTS AND METHODS: This was a retrospective study of outpatient oncology patients who received treatment at Moffitt Cancer Center. Press Ganey surveys were used to assess patient experience. Data from patients with appointments between April 1, 2020, and June 30, 2021, were analyzed. Patient experience was compared between telemedicine and in-person visits, and patient experience with telemedicine over time was described. RESULTS: A total of 33,318 patients reported Press Ganey data for in-person visits, and 5,950 reported Press Ganey data for telemedicine visits. Relative to patients with in-person visits, more patients with telemedicine visits gave higher satisfaction ratings for access (62.5% vs 75.8%, respectively) and care provider concern (84.2% vs 90.7%, respectively) (P<.001). When adjusted for age, race/ethnicity, sex, insurance, and clinic type, telemedicine visits consistently outperformed in-person visits over time regarding access and care provider concern (P<.001). There were no significant changes over time in satisfaction with telemedicine visits regarding access, care provider concern, telemedicine technology, or overall assessment (P>.05). CONCLUSIONS: In this study, a large oncology dataset showed that telemedicine resulted in better patient experience of care in terms of access and care provider concern compared with in-person visits. Patient experience of care with telemedicine visits did not change over time, suggesting that implementing telemedicine was effective.

Health-related quality of life and vulnerability among people with myelodysplastic syndromes: a US national study.

Health-related quality of life (HRQoL) and vulnerability are variably affected in patients with myelodysplastic syndromes (MDS) and other cytopenic states; however, the heterogeneity of these diseases has limited our understanding of these domains. The National Heart, Lung, and Blood Institute-sponsored MDS Natural History Study is a prospective cohort enrolling patients undergoing workup for suspected MDS in the setting of cytopenias. Untreated patients undergo bone marrow assessment with central histopathology review for assignment as MDS, MDS/myeloproliferative neoplasm (MPN), idiopathic cytopenia of undetermined significance (ICUS), acute myeloid leukemia (AML) with <30% blasts, or "At-Risk." HRQoL data are collected at enrollment, including the MDS-specific Quality of Life in Myelodysplasia Scale (QUALMS). Vulnerability is assessed with the Vulnerable Elders Survey. Baseline HRQoL scores from 449 patients with MDS, MDS/MPN, AML <30%, ICUS or At-Risk were similar among diagnoses. In MDS, HRQoL was worse for vulnerable participants (eg, mean Patent-Reported Outcomes Management Information Systems [PROMIS] Fatigue of 56.0 vs 49.5; P < .001) and those with worse prognosis (eg, mean Euroqol-5 Dimension-5 Level [EQ-5D-5L] of 73.4, 72.7, and 64.1 for low, intermediate, and high-risk disease; P = .005). Among vulnerable MDS participants, most had difficulty with prolonged physical activity (88%), such as walking a quarter mile (74%). These data suggest that cytopenias leading to MDS evaluation are associated with similar HRQoL, regardless of eventual diagnosis, but with worse HRQoL among the vulnerable. Among those with MDS, lower-risk disease was associated with better HRQoL, but the relationship was lost among the vulnerable, showing for the first time that vulnerability trumps disease risk in affecting HRQoL. This study is registered at www.clinicaltrials.gov as NCT02775383.

Demographic disparities in receipt of care at a comprehensive cancer center.

BACKGROUND: National Cancer Institute cancer centers (NCICCs) provide specialized cancer care including precision oncology and clinical treatment trials. While these centers can offer novel therapeutic options, less is known about when patients access these centers or at what timepoint in their disease course they receive specialized care. This is especially important since precision diagnostics and receipt of the optimal therapy upfront can impact patient outcomes and previous research suggests that access to these centers may vary by demographic characteristics. Here, we examine the timing of patients' presentation at Moffitt Cancer Center (MCC) relative to their initial diagnosis across several demographic characteristics. METHODS: A retrospective cohort study was conducted among patients who presented to MCC with breast, colon, lung, melanoma, and prostate cancers between December 2008 and April 2020. Patient demographic and clinical characteristics were obtained from the Moffitt Cancer Registry. The association between patient characteristics and the timing of patient presentation to MCC relative to the patient's cancer diagnosis was examined using logistic regression. RESULTS: Black patients (median days = 510) had a longer time between diagnosis and presentation to MCC compared to Whites (median days = 368). Black patients were also more likely to have received their initial cancer care outside of MCC compared to White patients (odds ratio [OR] and 95% confidence interval [CI] = 1.45 [1.32-1.60]). Furthermore, Hispanics were more likely to present to MCC at an advanced stage compared to non-Hispanic patients (OR [95% CI] = 1.28 [1.05-1.55]). CONCLUSIONS: We observed racial and ethnic differences in timing of receipt of care at MCC. Future studies should aim to identify contributing factors for the development of novel mitigation strategies and assess whether timing differences in referral to an NCICC correlate with long-term patient outcomes.

Assessing the Contribution of Scanned Outside Documents to the Completeness of Real-World Data Abstraction.

PURPOSE: Electronic health record (EHR) data are widely used in precision medicine, quality improvement, disease surveillance, and population health management. However, a significant amount of EHR data are stored in unstructured formats including scanned documents external to the treatment facility presenting an informatics challenge for secondary use. Studies are needed to characterize the clinical information uniquely available in scanned outside documents (SODs) to understand to what extent the availability of such information affects the use of these real-world data for cancer research. MATERIALS AND METHODS: Two independent EHR data abstractions capturing 30 variables commonly used in oncology research were conducted for 125 patients treated for advanced non-small-cell lung cancer at a comprehensive cancer center, with and without consideration of SODs. Completeness and concordance were compared between the two abstractions, overall, and by patient groups and variable types. RESULTS: The overall completeness of the data with SODs was 77.6% as compared with 54.3% for the abstraction without SODs. The differences in completeness were driven by data related to biomarker tests, which were more likely to be uniquely available in SODs. Such data were prone to missingness among patients who were diagnosed externally. CONCLUSION: There were no major differences in completeness between the two abstractions by demographics, diagnosis, disease progression, performance status, or oral therapy use. However, biomarker data were more likely to be uniquely contained in the SODs. Our findings may help cancer centers prioritize the types of SOD data being abstracted for research or other secondary purposes.

Advancing Digital Health Innovation in Oncology: Priorities for High-Value Digital Transformation in Cancer Care.

Although health care delivery is becoming increasingly digitized, driven by the pursuit of improved access, equity, efficiency, and effectiveness, progress does not appear to be equally distributed across therapeutic areas. Oncology is renowned for leading innovation in research and in care; digital pathology, digital radiology, real-world data, next-generation sequencing, patient-reported outcomes, and precision approaches driven by complex data and biomarkers are hallmarks of the field. However, remote patient monitoring, decentralized approaches to care and research, "hospital at home," and machine learning techniques have yet to be broadly deployed to improve cancer care. In response, the Digital Medicine Society and Moffitt Cancer Center convened a multistakeholder roundtable discussion to bring together leading experts in cancer care and digital innovation. This viewpoint highlights the findings from these discussions, in which experts agreed that digital innovation is lagging in oncology relative to other therapeutic areas. It reports that this lag is most likely attributed to poor articulation of the challenges in cancer care and research best suited to digital solutions, lack of incentives and support, and missing standardized infrastructure to implement digital innovations. It concludes with suggestions for actions needed to bring the promise of digitization to cancer care to improve lives.

Estimated Carbon Emissions Savings With Shifts From In-Person Visits to Telemedicine for Patients With Cancer.

Importance: While the health care community advocates broadly for climate change policy, medical professionals can look within care practices to assess their contribution to carbon dioxide (CO2) emissions, and provide solutions wherever possible. Telemedicine can help in mitigating climate change by providing care from a distance. Objective: To assess the carbon savings achieved from telemedicine visits. Design, Setting, and Participants: This cross-sectional study of telemedicine visits was conducted at a single-institution National Cancer Institute (NCI)-designated comprehensive cancer center. Eligible patients were aged 18 years and above, completed telemedicine visits from April 1, 2020, to June 20, 2021, and had a Florida mailing address documented in their electronic medical record. Groups were divided between those within driving time of 60 minutes (1-way) to the cancer center vs those living beyond 60 minutes of drive time. Data were analyzed between April 2020 and June 2021. Main Outcomes and Measures: Carbon emission savings from telemedicine, measured in total and average per-visit savings. Results: A total 49 329 telemedicine visits with 23 228 patients were conducted from April 1, 2020, to June 30, 2021. A total 21 489 visits were for patients with driving time of 60 minutes or less (median [IQR] age, 62.0 [52.0-71.0] years; 12 334 [57.4%] female; 1685 [7.8%] Black, 1500 [7.0%] Hispanic, 16 010 [74.5%] non-Hispanic White), while 27 840 visits were for patients with driving time greater than 60 minutes (median [IQR] age, 67.0 [57.0-74.0] years; 14 372 [51.6%] female; 1056 [3.8%] Black, 1364 [5.0%] Hispanic, 22 457 [80.7%] non-Hispanic White). For patients living within a driving time of 60 minutes from the cancer center, 424 471 kg CO2 emissions were saved (mean [SD] emissions savings, 19.8 [9.4] kg CO2 per visit) due to telemedicine-equivalent to 91.5 passenger vehicles driven for 1 year. For patients whose driving distance was greater than 60 minutes, 2 744 248 kg CO2 emissions were saved (mean emissions savings, 98.6 [54.8] kg CO2 per visit)-equivalent to 591 passenger vehicles driven for 1 year. Conclusions and Relevance: Using a large data set, this cross-sectional analysis highlighted the carbon emissions savings due to telemedicine in oncology. This has important implications in reducing health care-related carbon footprint.

Oncology Providers' and Professionals' Experiences With Suicide Risk Screening Among Patients With Head and Neck Cancer: A Qualitative Study.

PURPOSE: There has been limited study of the implementation of suicide risk screening for patients with head and neck cancer (HNC) as a part of routine care. To address this gap, this study assessed oncology providers' and professionals' perspectives about barriers and facilitators of implementing a suicide risk screening among patients with HNC. MATERIALS AND METHODS: All patients with HNC with an in-person visit completed a suicide risk screening on an electronic tablet. Patients reporting passive death wish were then screened for active suicidal ideation and referred for appropriate intervention. Interviews were conducted with 25 oncology providers and professionals who played a key role in implementation including nurses, medical assistants, patient access representatives, advanced practice providers, physicians, social workers, and informatics staff. The interview guide was based on the Consolidated Framework for Implementation Research. Interviews were transcribed and analyzed for themes. RESULTS: Participants identified multilevel implementation barriers, such as intervention level (eg, patient difficulty with using a tablet), process level (eg, limited nursing engagement), organizational level (eg, limited clinic Wi-Fi connectivity), and individual level (eg, low clinician self-efficacy for interpreting and acting upon patient-reported outcome scores). Participants noted facilitators, such as effective care coordination across nursing and social work staff and the opportunity for patients to be screened multiple times. Participants recommended strengthening patient and clinician education and providing patients with other modalities for data entry (eg, desktop computer in the waiting room). CONCLUSION: Participants identified important intervention modifications that may be needed to optimize suicide risk screening in cancer care settings.

U2AF1 and EZH2 mutations are associated with nonimmune hemolytic anemia in myelodysplastic syndromes.

Hemolysis is a well-recognized but poorly characterized phenomenon in a subset of patients with myelodysplastic syndromes (MDS). Its pathobiological basis seems to underpin a nonimmune etiology whose clinical significance has not been adequately characterized. Hemolysis in MDS is often attributed to either ineffective intramedullary erythropoiesis or acquired hemoglobinopathies and red blood cell (RBC) membrane defects. These heterogeneous processes have not been associated with specific genetic subsets of the disease. We aimed to describe the prevalence of hemolysis among patients with MDS, their baseline characteristics, molecular features, and resulting impact on outcomes. We considered baseline serum haptoglobin <10 mg/dL a surrogate marker for intravascular hemolysis. Among 519 patients, 10% had hemolysis. The baseline characteristics were similar among both groups. Only 13% of patients with hemolysis were Coombs-positive, suggesting that hemolysis in MDS is largely not immune-mediated. Inferior survival trends were observed among lower-risk patients with MDS undergoing hemolysis. Decreased response rates to erythropoiesis-stimulating agents (ESA) and higher responses to hypomethylating agents (HMA) were also observed in the hemolysis group. U2AF1 and EZH2 hotspot mutations were more prevalent among those undergoing hemolysis (P < .05). U2AF1 mutations were observed in 30% of patients with hemolysis and occurred almost exclusively at the S34 hotspot. Somatic mutations encoding splicing factors may affect erythrocyte membrane components, biochemical properties, and RBC metabolic function, which underpin the development of atypical clones from erythroid precursors in MDS presenting with hemolysis. Future studies will explore the contribution of altered splicing to the development of acquired hemoglobinopathies.

History of keratinocyte carcinoma and survival after a second primary malignancy: the Moffitt Cancer Center patient experience.

PURPOSE: History of keratinocyte carcinoma (KC) has been associated with survival following the diagnosis of a second primary malignancy (SPM), with the direction of the association varying by cancer type. Research is needed to elucidate the role of other key factors in this association. METHODS: A retrospective cohort study was conducted among patients newly diagnosed and/or treated at Moffitt Cancer Center in December 2008-April 2020 with breast cancer, lung cancer, melanoma, colon cancer, prostate cancer, and non-Hodgkin lymphoma/chronic lymphocytic leukemia (NHL/CLL) (n = 29,156). History of KC was obtained from new patient intake questionnaires. Age- and stage-adjusted hazard ratios (HR) and 95% confidence intervals (CI) were calculated to estimate the association between history of KC and survival following each cancer, stratified by demographic/clinical characteristics. RESULTS: KC history was most prevalent in patients with melanoma (28.7%), CLL (19.8%) and lung cancer (16.1%). KC history was associated with better overall survival following prostate cancer (HR = 0.74, 95% CI = 0.55-0.99) and poorer overall survival following CLL (HR = 1.73, 95% CI = 1.10-2.71). Patients with a history of KC experienced better survival within the first four years of a melanoma diagnosis (HR = 0.79, 95% CI = 0.67-0.92); whereas poorer survival was observed for patients who survived 7 + years after a melanoma diagnosis (HR = 2.18, 95% CI = 1.17-4.05). Stratification by treatment and stage revealed directional differences in the associations between KC history and survival among patients with breast cancer and melanoma. CONCLUSIONS: KC history may be a predictor of survival following an SPM, possibly serving as a marker of immune function and/or DNA damage repair capacity.

A Question-and-Answer System to Extract Data From Free-Text Oncological Pathology Reports (CancerBERT Network): Development Study.

BACKGROUND: Information in pathology reports is critical for cancer care. Natural language processing (NLP) systems used to extract information from pathology reports are often narrow in scope or require extensive tuning. Consequently, there is growing interest in automated deep learning approaches. A powerful new NLP algorithm, bidirectional encoder representations from transformers (BERT), was published in late 2018. BERT set new performance standards on tasks as diverse as question answering, named entity recognition, speech recognition, and more. OBJECTIVE: The aim of this study is to develop a BERT-based system to automatically extract detailed tumor site and histology information from free-text oncological pathology reports. METHODS: We pursued three specific aims: extract accurate tumor site and histology descriptions from free-text pathology reports, accommodate the diverse terminology used to indicate the same pathology, and provide accurate standardized tumor site and histology codes for use by downstream applications. We first trained a base language model to comprehend the technical language in pathology reports. This involved unsupervised learning on a training corpus of 275,605 electronic pathology reports from 164,531 unique patients that included 121 million words. Next, we trained a question-and-answer (Q&A) model that connects a Q&A layer to the base pathology language model to answer pathology questions. Our Q&A system was designed to search for the answers to two predefined questions in each pathology report: What organ contains the tumor? and What is the kind of tumor or carcinoma? This involved supervised training on 8197 pathology reports, each with ground truth answers to these 2 questions determined by certified tumor registrars. The data set included 214 tumor sites and 193 histologies. The tumor site and histology phrases extracted by the Q&A model were used to predict International Classification of Diseases for Oncology, Third Edition (ICD-O-3), site and histology codes. This involved fine-tuning two additional BERT models: one to predict site codes and another to predict histology codes. Our final system includes a network of 3 BERT-based models. We call this CancerBERT network (caBERTnet). We evaluated caBERTnet using a sequestered test data set of 2050 pathology reports with ground truth answers determined by certified tumor registrars. RESULTS: caBERTnet's accuracies for predicting group-level site and histology codes were 93.53% (1895/2026) and 97.6% (1993/2042), respectively. The top 5 accuracies for predicting fine-grained ICD-O-3 site and histology codes with 5 or more samples each in the training data set were 92.95% (1794/1930) and 96.01% (1853/1930), respectively. CONCLUSIONS: We have developed an NLP system that outperforms existing algorithms at predicting ICD-O-3 codes across an extensive range of tumor sites and histologies. Our new system could help reduce treatment delays, increase enrollment in clinical trials of new therapies, and improve patient outcomes.

Effect of prior antibiotic or chemotherapy treatment on immunotherapy response in non-small cell lung cancer.

BACKGROUND: Treatment outcomes of advanced non-small cell lung cancer (NSCLC) have substantially improved with immune checkpoint inhibitors (ICI), although only approximately 19% of patients respond to immunotherapy alone, increasing to 58% with the addition of chemotherapy. The gut microbiome has been recognized as a modulator of ICI response via its priming effect on the host immune response. Antibiotics as well as chemotherapy reduce gut microbial diversity, hence altering composition and function of the gut microbiome. Since the gut microbiome may modify ICI efficacy, we conducted a retrospective study evaluating the effects of prior antibiotic or chemotherapy use on NSCLC patient response to ICI. METHODS: We retrospectively evaluated 256 NSCLC patients treated between 2011-2017 at Moffitt Cancer Center with ICI ± chemotherapy, examining the associations between prior antibiotic or chemotherapy use, overall response rate and survival. Relative risk regression using a log-link with combinatorial expectation maximization algorithm was performed to analyze differences in response between patients treated with antibiotics or chemotherapy versus patients who didn't receive antibiotics or chemotherapy. Cox proportional hazards models were constructed to evaluate associations between risk factors and overall survival. RESULTS: Only 46 (18% of 256) patients used antibiotics prior to and/or during ICI treatment, and 146 (57%) had prior chemotherapy. Antibiotic users were 8% more likely to have worse overall response rate (RR:1.08; CI:0.93-1.26; p = 0.321), as well as a 35% worse overall survival (HR:1.35; CI:0.91-2.02; p = 0.145), although results were not statistically significant. However, prior use of chemotherapy was significantly associated with poor ICI response (RR:1.24; CI:1.05-1.47; p = 0.013) and worse overall survival (HR:1.47; CI:1.07-2.03; p = 0.018). CONCLUSIONS: Patients receiving antibiotics prior to and/or during ICI therapy might experience worse treatment outcomes and survival than unexposed patients, although these associations were not statistically significant and hence warrant further prospective study. Prior chemotherapy significantly reduced ICI response and overall survival. Antibiotic or chemotherapy exposure may negatively impact ICI response, perhaps through disruption of the eubiotic gut microbiome.

Natural History of Incident and Persistent Cutaneous Human Papillomavirus and Human Polyomavirus Infections.

BACKGROUND: Cutaneous human papillomaviruses (cuHPV) and polyomaviruses (HPyV) have been implicated in skin cancers; however, interpretation of findings across studies is complicated by limited understanding of the natural history of these infections across normal tissue types. METHODS: In total, 675 eyebrow hair (EBH) and skin swab (SSW) samples were collected from 71 skin cancer screening patients every 6 months over 2 years and measured for presence of ß-HPV, γ-HPV, and HPyV. Incidence, persistence, and clearance of cuHPV/HPyV were estimated, and risk factors associated with infection were examined. RESULTS: Prevalence, incidence, and persistence of ß-HPV, γ-HPV, and HPyV were consistently higher in SSW than in EBH, with types 5, 24, 49, 76 and Merkel cell polyomavirus (MCPyV) having incidence rates greater than 20 per 1000 person-months. Prevalent γ-HPV EBH infections persisted more often in women (P = .024), incident ß-HPV EBH infections persisted less often among individuals with history of blistering sunburn (P = .019), and prevalent MCPyV SSW infections persisted more often in those with a history of skin cancer (P = .033). CONCLUSIONS: Incidence and persistence of cuHPV/HPyV were observed in SSW and EBH; however, none of the risk factors examined were commonly associated with cuHPV/HPyV infections across normal tissue types.

Longitudinal patient-reported outcomes and survival among early-stage non-small cell lung cancer patients receiving stereotactic body radiotherapy.

BACKGROUND AND PURPOSE: The study objective was to determine whether longitudinal changes in patient-reported outcomes (PROs) were associated with survival among early-stage, non-small cell lung cancer (NSCLC) patients undergoing stereotactic body radiation therapy (SBRT). MATERIALS AND METHODS: Data were obtained from January 2015 through March 2020. We ran a joint probability model to assess the relationship between time-to-death, and longitudinal PRO measurements. PROs were measured through the Edmonton Symptom Assessment Scale (ESAS). We controlled for other covariates likely to affect symptom burden and survival including stage, tumor diameter, comorbidities, gender, race/ethnicity, relationship status, age, and smoking status. RESULTS: The sample included 510 early-stage NSCLC patients undergoing SBRT. The median age was 73.8 (range: 46.3-94.6). The survival component of the joint model demonstrates that longitudinal changes in ESAS scores are significantly associated with worse survival (HR: 1.04; 95% CI: 1.02-1.05). This finding suggests a one-unit increase in ESAS score increased probability of death by 4%. Other factors significantly associated with worse survival included older age (HR: 1.04; 95% CI: 1.03-1.05), larger tumor diameter (HR: 1.21; 95% CI: 1.01-1.46), male gender (HR: 1.87; 95% CI: 1.36-2.57), and current smoking status (HR: 2.39; 95% CI: 1.25-4.56). CONCLUSION: PROs are increasingly being collected as a part of routine care delivery to improve symptom management. Healthcare systems can integrate these data with other real-world data to predict patient outcomes, such as survival. Capturing longitudinal PROs-in addition to PROs at diagnosis-may add prognostic value for estimating survival among early-stage NSCLC patients undergoing SBRT.