Utilizing SEER Cancer Registries for Population-Based Cancer Survivor Epidemiologic Studies: A Feasibility Study

BACKGROUND

While the primary role of central cancer registries in the United States is to provide vital information needed for cancer surveillance and control, these registries can also be leveraged for population-based epidemiologic studies of cancer survivors. This study was undertaken to assess the feasibility of using the NCI's Surveillance, Epidemiology, and End Results (SEER) Program registries to rapidly identify, recruit, and enroll individuals for survivor research studies and to assess their willingness to engage in a variety of research activities.

METHODS

In 2016 and 2017, six SEER registries recruited both recently diagnosed and longer-term survivors with early age-onset multiple myeloma or colorectal, breast, prostate, or ovarian cancer. Potential participants were asked to complete a survey, providing data on demographics, health, and their willingness to participate in various aspects of research studies.

RESULTS

Response rates across the registries ranged from 24.9% to 46.9%, with sample sizes of 115 to 239 enrolled by each registry over a 12- to 18-month period. Among the 992 total respondents, 90% answered that they would be willing to fill out a survey for a future research study, 91% reported that they would donate a biospecimen of some type, and approximately 82% reported that they would consent to have their medical records accessed for research.

CONCLUSIONS

This study demonstrated the feasibility of leveraging SEER registries to recruit a geographically and racially diverse group of cancer survivors.

IMPACT

Central cancer registries are a source of high-quality data that can be utilized to conduct population-based cancer survivor studies.

Population-Based Study on Cancer Subtypes, Guideline-Concordant Adjuvant Therapy, and Survival Among Women With Stage I-III Breast Cancer

BACKGROUND

Breast cancer subtype is a key determinant in treatment decision-making, and also effects survival outcome. In this population-based study, in-depth analyses were performed to examine the impact that breast cancer subtype and receipt of guideline-concordant adjuvant systemic therapy (AST) have on survival using a population-based cancer registry's data.

METHODS

Women aged ≥20 years with microscopically confirmed stage I-III breast cancer diagnosed in 2011 were identified from the Louisiana Tumor Registry. Breast cancer subtypes were categorized based on hormone receptor (HR) and HER2 status. Guideline-concordant treatment was defined using the NCCN Guidelines for Breast Cancer. Logistic regression was applied to identify factors associated with guideline-concordant AST receipt. Kaplan-Meier survival curves were generated to compare survival among subtypes by AST receipt status, and a semiparametric additive hazard model was used to verify the factors impacting survival outcome.

RESULTS

Of 2,214 eligible patients, most (70.8%) were HR+/HER2- followed by HR-/HER2- (14.4%), and 78.6% received guideline-concordant AST. Compared with patients with the HR+/HER2+ subtype, women with other subtypes were more likely to be guideline-concordant after adjusting for sociodemographic and clinical variables. Women with the HR-/HER2+ or HR-/HER2- subtype had a higher risk of any-cause and breast cancer-specific death than those with the HR+/HER2+ subtype. Those who did not receive AST had an additional adjusted hazard of 0.0191 (P=.0001) in overall survival and 0.0126 (P=.0011) in cause-specific survival compared with those who received AST.

CONCLUSIONS

Most patients received guideline-concordant AST, except for those with the HR+/HER2+ subtype. Patients receiving guideline-adherent adjuvant therapy had better survival outcomes across all breast cancer subtypes.

Abstract 224: Evaluation of next generation sequencing of DNA and RNA from archival formalin-fixed, paraffin-embedded pancreatic cancer tissue: A pilot study of the SEER-linked virtual tissue repository

As formalin-fixed, paraffin-embedded (FFPE) tissue is being utilized for next-generation sequencing (NGS) in research and clinical settings, we conducted a study through the Surveillance, Epidemiology and End Results (SEER)-Linked Virtual Tissue Repository (VTR) Pilot Program to determine the quality of sequencing data obtained using FFPE-derived DNA and RNA. Forty-eight pancreatic ductal adenocarcinoma (PDAC) patients, comprising 24 case-control pairs based on survival time (≥5 years [cases] vs <24 months [controls]), were selected. Participating SEER registries obtained selected diagnostic tissue blocks collected clinically and stored for 4-18 years. DNA and RNA were extracted from the FFPE specimens for 36 patients (18 pairs). Whole genome (WGS) and whole exome sequencing (WES) were performed on tumor and normal DNA from 16 patients, and a methylation array was conducted on tumor DNA from 6 of these patients. RNA-Seq was conducted on tumor RNA from 36 patients. The median coverage depths for tumor were above 300x for WES and 60x for WGS. However, the majority of sequencing reads (>60%) were duplicates. Concordant mutations (SNVs, MNVs and indels) were >50% by WGS and WES from the majority of samples (n=11, 69%), and the most common discordant mutations were C>T. On average, mutant allele frequencies (MAFs) were 20% in coding regions and 15% across the whole genome, consistent with tumor content as measured by methylation analysis for five tumor samples (18%, 22%, 28%, 42%, and 50%). WES and/or WGS revealed that specimens for five of 27 PDAC subjects tested had a high fraction of variants overlapping with germline variants in dbSNP (≥20%), indicating that tumor cellularity was low among these samples. TP53, KRAS, CDKN2A, SMAD4, and RNF43 were the most frequently mutated genes from these specimens, consistent with genes reported in studies using fresh frozen tissue. Point mutations comprised most of the gene variants, and indels were found in CDKN2A, SMAD4, and RNF43. Most of the mutation status (e.g. missense, nonsense or indels) were concordantly called by WES and WGS (e.g., 81% for TP53, 100% CDKN2A, 94% SMAD4, and 94% RNF43). Most discordant calls were mutations identified by WES but not WGS (e.g., 8 [50%] for KRAS and 3 [19%] TP53). All samples yielded RNA-Seq reads with <30% exonic mapping, 39% (14) of which had <10% exonic mapping. Our study provided important evidence for NGS applications on DNA and RNA from archival PDAC FFPE tissue specimens stored for up to 18 years. These findings demonstrate that, with sufficient tumor content and coverage depth, FFPE-derived DNA is adequate for identifying somatic driver gene mutations in PDAC patients and that the it is feasible to utilize the population-based, SEER-Linked VTR as an infrastructure for obtaining diagnostic tissue for molecular studies.

Citation Format: Yao Yuan, Alison Van Dyke, Valentina Petkov, Alyssa Tuan, Aatur Singhi, Lynn Matrisian, Lola Rahib, John Pearson, Katia Nones, Nicola Waddell, Yongmei Zhao, Tsai-wei Shen, Bao Tran, Jyoti Shetty, Elizabeth Gillanders, Danielle Carrick, Rosemary Cress, Lloyd Mueller, Brenda Hernandez, Charles Lynch, Thomas Tucker, Xiao-Cheng Wu, Lynne Penberthy. Evaluation of next generation sequencing of DNA and RNA from archival formalin-fixed, paraffin-embedded pancreatic cancer tissue: A pilot study of the SEER-linked virtual tissue repository [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 224.

Radiotherapy after radical prostatectomy: Effect of timing of postprostatectomy radiation on functional outcomes

INTRODUCTION AND OBJECTIVE

The timing of radiotherapy (RT) after prostatectomy is controversial, and its effect on sexual, urinary, and bowel function is unknown. This study seeks to compare patient-reported functional outcomes after radical prostatectomy (RP) and postprostatectomy radiation as well as elucidate the timing of radiation to allow optimal recovery of function.

METHODS

The Comparative Effectiveness Analysis of Surgery and Radiation (CEASAR) study is a prospective, population-based, observational study of men with localized prostate cancer. Patient-reported sexual, urinary, and bowel functional outcomes were measured using the 26-item Expanded Prostate Index Composite at baseline and at 6, 12, 36, and 60 months after enrollment. Functional outcomes were compared among men undergoing RP alone, post-RP adjuvant radiation (RP + aRT), and post-RP salvage radiation (RP + sRT) using multivariable models controlling for baseline clinical, demographic, and functional characteristics.

RESULTS

Among 1,482 CEASAR participants initially treated with RP for clinically localized prostate cancer, 11.5% (N = 170) received adjuvant (aRT, N = 57) or salvage (sRT, N = 113) radiation. Men who received post-RP RT had worse scores in all domains (sexual function [-9.0, 95% confidence interval {-14.5, -3.6}, P < 0.001], incontinence [-8.8, {-14.0, -3.6}, P < 0.001], irritative voiding [-5.9, {-9.0, -2.8}, P < 0.001], bowel irritative [-3.5, {-5.8, -1.2}, P = 0.002], and hormonal function [-4.5, {-7.2, -1.7}, P = 0.001]) compared to RP alone at 5 years of follow-up. Compared to men treated with RP alone in an adjusted linear model, sRT was associated with significantly worse scores in all functional domains. aRT was associated with significantly worse incontinence, urinary irritation, and hormonal function domain scores compared to RP alone at 5 years of follow-up. On multivariable modeling, RT administered approximately 24 months after RP was associated with the smallest decline in sexual domain score, with an adjusted mean decrease of 8.85 points (95% confidence interval [-19.8, 2.1]) from post-RP, pre-RT baseline.

CONCLUSIONS

In men with localized prostate cancer, post-RP RT was associated with significantly worse sexual, urinary, and bowel function domain scores at 5 years compared to RP alone. Radiation delayed for approximately 24 months after RP may be optimal for preserving erectile function compared to radiation administered closer to the time of RP.

Knowledge Graph-Enabled Cancer Data Analytics

Cancer registries collect unstructured and structured cancer data for surveillance purposes which provide important insights regarding cancer characteristics, treatments, and outcomes. Cancer registry data typically (1) categorize each reportable cancer case or tumor at the time of diagnosis, (2) contain demographic information about the patient such as age, gender, and location at time of diagnosis, (3) include planned and completed primary treatment information, and (4) may contain survival outcomes. As structured data is being extracted from various unstructured sources, such as pathology reports, radiology reports, medical records, and stored for reporting and other needs, the associated information representing a reportable cancer is constantly expanding and evolving. While some popular analytic approaches including SEER*Stat and SAS exist, we provide a knowledge graph approach to organizing cancer registry data. Our approach offers unique advantages for timely data analysis and presentation and visualization of valuable information. This knowledge graph approach semantically enriches the data, and easily enables linking with third-party data which can help explain variation in cancer incidence patterns, disparities, and outcomes. We developed a prototype knowledge graph based on the Louisiana Tumor Registry dataset. We present the advantages of the knowledge graph approach by examining: i) scenario-specific queries, ii) links with openly available external datasets, iii) schema evolution for iterative analysis, and iv) data visualization. Our results demonstrate that this graph based solution can perform complex queries, improve query run-time performance by up to 76%, and more easily conduct iterative analyses to enhance researchers' understanding of cancer registry data.

Prevalence of Eustachian Tube Dysfunction in the US Elderly Population

Objective

To investigate the prevalence of eustachian tube dysfunction (ETD) in elderly adults in the United States and its association with other upper aerodigestive inflammatory processes.

Study Design

Cross-sectional study.

Setting

Population based.

Subjects and Methods

In total, 147,805 patients without malignancy were compared to 13,804 demographically matched patients with malignancy of the upper aerodigestive tract (UADT) by querying the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked database for patients aged 66 to 99 years between 2003 and 2011. The prevalence of ETD and inflammatory diseases among these patients was compared. Association between ETD, other upper aerodigestive inflammatory processes, and UADT malignancies was evaluated.

Results

The prevalence of ETD was 5.44% among patients without malignancy and 9.08% in those with cancer (odds ratio [OR], 1.73; 95% CI, 1.63-1.84). Patients with ETD in the control population were more likely (OR, 95% CI) to be diagnosed with chronic rhinitis (5.00, 4.70-5.33), chronic sinusitis (4.20, 3.98-4.43), allergic rhinitis (4.27, 4.08-4.47), and gastroesophageal reflux disease (GERD) (2.42, 2.31-2.53). Patients with ETD and chronic rhinitis (1.43, 1.24-1.65), chronic sinusitis (1.57, 1.38-1.78), and acute otitis media (1.33, 1.08-1.65) were associated with higher rates of UADT malignancy.

Conclusion

Over 5% of patients older than 65 in the United States are diagnosed with ETD in the absence of UADT malignancy. Associations between ETD and chronic rhinitis, chronic sinusitis, allergic rhinitis, and GERD in the absence of UADT malignancy suggest that some patients may benefit from treatment of inflammatory disease as a cause of ETD.

Abstract IA08: Exploring social determinants of cancer outcomes using multilevel spatial analysis

Health disparities research increasingly relies on a social determinants of health approach, which focuses on identifying modifiable social and environmental factors that adversely affect health outcomes to create health disparities. We have used a multilevel framework that leverages cancer registry and US census data to examine the role of neighborhood social determinants in cancer incidence and survival in Louisiana. Multilevel studies of cases of colorectal, liver, and breast cancer reveal that differential exposure to unfavorable social and physical environments do contribute to racial disparities in cancer. In particular, we have observed that concentrated disadvantage, a robust measure of neighborhood environment, was positively associated with the incidence of colorectal cancer and hepatocellular carcinoma but not triple-negative breast cancer.

Citation Format: Richard Scribner, Denise Danos, Claudia Leonardi, Tekeda Ferguson, Qingzhao Yu, Neal Simonsen, Xiao-Cheng Wu. Exploring social determinants of cancer outcomes using multilevel spatial analysis [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr IA08.

Abstract IA22: A comprehensive and integrated approach to identify factors associated with aggressive prostate cancer in African-Americans: The RESPOND Study

African American (AA) men have a &gt;60% higher incidence, are more likely to be diagnosed with aggressive and potentially lethal PCa, and are more than twice as likely to die from PCa than white (WH) men. Reasons for the greater burden of aggressive disease in AA men are unknown but are likely to include a multitude of factors including social factors such as lifetime stress, inherited susceptibility, and tumor-related features such as somatic alterations and local inflammation in the microenvironment. RESPOND (Research on Prostate Cancer in Men of African Ancestry: Defining the Roles of Genetics, Immunity and Social Stress) has been funded to study the reasons for increased risk of aggressive disease among AA and will include a comprehensive approach including the role of stress and the contextual environment (Project 1), germline susceptibility (Project 2), tumor somatic genetics (Project 3), inflammatory tumor microenvironment (Project 4), and, the integrated synergistic effects of these factors. Project 1: Social stress: AA men are uniquely exposed to high levels of social adversity such as discrimination, violence, crime, financial strain, and residence in poor-resourced environments, over their lifetime. These social stressors are experienced at multiple levels, including individual, neighborhood, and institutional, ultimately leading to chronic and long-term stress. Social stressors may be a contributor to the development of aggressive PCa and high mortality. Project 2: Germline genetic factors: There is evidence to suggest genetic differences in the allelic architecture of PCa across populations. We found a region of the genome (8q24) that harbored genetic risk alleles that may contribute to the greater risk of PCa in AA men. Substantially larger collections of AA cases are needed to reveal a common susceptibility locus that is specific to high-risk disease. We will be expanding genome-wide efforts to identify susceptibility alleles for aggressive PCa in AA men. Project 3: Somatic genetic factors. PCa is a heterogeneous and multifocal disease with a diverse genetic component. Few genetic or molecular drivers of aggressive disease have been identified, and studies in AA men are critically limited. This is due to the fact that the majority of PCa profiling studies have focused on localized disease and largely on men of European ancestry. We will address this issue through the comprehensive and integrated genomic and transcriptomic analysis of a large number of clinically annotated aggressive vs. nonaggressive PCas in AA men. We anticipate that this study will identify genomic markers of aggressive PCa in general and AA PCa in particular, with some of these somatic events being therapeutically actionable, leading to new treatment paradigms for this lethal disease. Project 4: Inflammatory tumor microenvironment. There is evidence to suggest that inflammation drives the formation of precursor lesions to PCa development and may contribute to PCa progression. The consistent observation of overexpression of genes involved in inflammatory pathways in PCa tumors from AA men points to a proinflammatory immune cell component in the tumors of these men that may contribute to PCa progression and their higher PCa mortality. Integration of social and genetic information: Independently evaluating germline, somatic genome, and tumor microenvironment characteristics of PCa ignores the potential for shared biologic mechanism(s). The integration of data across these domains will lead to the identification of shared genes and/or pathways that define more homogeneous, and clinically important, subsets of PCa tumors. The vast genomic and molecular data generated in the same individuals will facilitate the estimation of tumor subgroups as a function of somatic and tumor microenvironment and an assessment of their relationship with other potential PCa risk factors, such as socioeconomic factors, lifetime stress, and genomic germline variation. We will evaluate the association of integrated molecular profiles with recurrence and survival in the future. The RESPOND Cohort: We are in the process of recruiting a cohort of 10,000 incident AA PCa cases to support the four research projects. The men, diagnosed between 2015-2018, are being recruited through cancer registries in 7 states (California, Florida, Georgia, Louisiana, Michigan, New Jersey, and Texas). The fieldwork procedures include enrolling the men by completion of a mailed (or online) survey, followed by a request to obtain a saliva sample and HIPAA authorization to obtain tumor tissue. Of the ~23,000 AA PCa patients we will contact, we estimate to receive a survey from 45% (10,050), DNA from 33% (7,543), HIPAA release from 26% (6,030) and tumor samples for 13% (3,015). Impact: In Project 1, we are studying the impact of multilevel stressors over the lifecourse on risk of developing aggressive and lethal PCa, providing both a means of identifying high-risk men for targeted prevention and treatment, as well as informing the etiology by which tumor aggressiveness arises. In Project 2, the ability to better understand, based on inherited variation, which AA men are at greater risk of developing aggressive and lethal PCa will help in the development of targeted screening and prevention strategies. In Project 3, delineating the genomic events that are acquired during the development of PCa in AA men may guide the development of targeted therapeutic strategies for men whose tumors display aggressive molecular signatures. In Project 4, the ability to define immune profiles associated with aggressive PCa in AA men could inform the development of cancer prevention and/or treatment strategies. Altogether, this body of research will provide impactful information as to the underlying factors that contribute to aggressive PCa in AA men.

Citation Format: Ann S. Hamilton, Scarlett Gomez, Xiao-Cheng Wu, Kevin Ward, Melissa Bondy, Rosemary Cress, Jennifer Beebe-Dimmer, Karen Pawlish, Jong Park, Iona Cheng, Antoinette Stroup, Thomas Sellers, Susan Gundell, Angelo Demarzo, Denise Modjeski, Stephen Chanock, Salma Shariff-Marco, Mindy DeRouen, John Carpten, Franklin Huang, Karen Sfanos, Tamara Lotan, David Conti, Christopher Haiman. A comprehensive and integrated approach to identify factors associated with aggressive prostate cancer in African-Americans: The RESPOND Study [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr IA22.

Using case-level context to classify cancer pathology reports

Individual electronic health records (EHRs) and clinical reports are often part of a larger sequence-for example, a single patient may generate multiple reports over the trajectory of a disease. In applications such as cancer pathology reports, it is necessary not only to extract information from individual reports, but also to capture aggregate information regarding the entire cancer case based off case-level context from all reports in the sequence. In this paper, we introduce a simple modular add-on for capturing case-level context that is designed to be compatible with most existing deep learning architectures for text classification on individual reports. We test our approach on a corpus of 431,433 cancer pathology reports, and we show that incorporating case-level context significantly boosts classification accuracy across six classification tasks-site, subsite, laterality, histology, behavior, and grade. We expect that with minimal modifications, our add-on can be applied towards a wide range of other clinical text-based tasks.

An Evaluation of the Utility of Big Data to Supplement Cancer Treatment Information: Linkage Between IQVIA Pharmacy Database and the Surveillance, Epidemiology, and End Results Program

Oral anticancer medications (OAMs) are increasingly utilized. We evaluated the representativeness and completeness of IQVIA, a large aggregator of pharmacy data, for breast cancer, colon cancer, chronic myeloid leukemia, and myeloma cases diagnosed in six Surveillance, Epidemiology, and End Results Program (SEER) registries between 2007 and 2011. Patient's SEER and SEER-Medicare data were linked and compared with IQVIA pharmacy data from 2006 to 2012 for specific OAMs. Overall, 67.6% of SEER cases had a pharmacy claim in IQVIA during the treatment assessment window. This varied by location, race and ethnicity, and insurance status. IQVIA consistently identified fewer cases who received an OAM of interest than SEER-Medicare. The difference was least pronounced for breast cancer agents and most pronounced for myeloma agents. The IQVIA pharmacy database included a large portion of persons in the SEER areas. Future studies should assess receipt of OAMs for other cancer sites and in different SEER registries.

Development and Internal Validation of a Web-based Tool to Predict Sexual, Urinary, and Bowel Function Longitudinally After Radiation Therapy, Surgery, or Observation

BACKGROUND

Shared decision making to guide treatment of localized prostate cancer requires delivery of the anticipated quality of life (QOL) outcomes of contemporary treatment options (including radical prostatectomy [RP], intensity-modulated radiation therapy [RT], and active surveillance [AS]). Predicting these QOL outcomes based on personalized features is necessary.

OBJECTIVE

To create an easy-to-use tool to predict personalized sexual, urinary, bowel, and hormonal function outcomes after RP, RT, and AS.

DESIGN, SETTING, AND PARTICIPANTS

A prospective, population-based cohort study was conducted utilizing US cancer registries of 2563 men diagnosed with localized prostate cancer in 2011-2012.

INTERVENTION

Patient-reported urinary, sexual, and bowel function up to 5 yr after treatment.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Patient-reported urinary, sexual, bowel, and hormonal function through 5 yr after treatment were collected using the 26-item Expanded Prostate Index Composite (EPIC-26) questionnaire. Comprehensive models to predict domain scores were fit, which included age, race, D'Amico classification, body mass index, EPIC-26 baseline function, treatment, and standardized scores measuring comorbidity, general QOL, and psychosocial health. We reduced these models by removing the instrument scores and replacing D'Amico classification with prostate-specific antigen (PSA) and Gleason score. For the final model, we performed bootstrap internal validation to assess model calibration from which an easy-to-use web-based tool was developed.

RESULTS AND LIMITATIONS

The prediction models achieved bias-corrected R-squared values of 0.386, 0.232, 0.183, 0.214, and 0.309 for sexual function, urinary incontinence, urinary irritative, bowel, and hormonal domains, respectively. Differences in R-squared values between the comprehensive and parsimonious models were small in magnitude. Calibration was excellent. The web-based tool is available at https://statez.shinyapps.io/PCDSPred/.

CONCLUSIONS

Functional outcomes after treatment for localized prostate cancer can be predicted at the time of diagnosis based on age, race, PSA, biopsy grade, baseline function, and a general question regarding overall health. Providers and patients can use this prediction tool to inform shared decision making.

PATIENT SUMMARY

In this report, we studied patient-reported sexual, urinary, hormonal, and bowel function through 5 yr after treatment with radical prostatectomy, radiation therapy, or active surveillance for localized prostate cancer. We developed a web-based predictive tool that can be used to predict one's outcomes after treatment based on age, race, prostate-specific antigen, biopsy grade, pretreatment baseline function, and a general question regarding overall health. We hope both patients and providers can use this tool to better understand expected outcomes after treatment, further enhancing shared decision making between providers and patients.

Trend and survival benefit of Oncotype DX use among female hormone receptor-positive breast cancer patients in 17 SEER registries, 2004-2015

PURPOSE

To examine (1) the trend and associated factors of Oncotype DX (ODX) use among hormone receptor-positive (HR+) breast cancer (BC) patients in 2004-2015; (2) the trend of reported chemotherapy by Recurrence Score (RS); and (3) the survival differences associated with ODX use.

METHODS

ODX data from Genomic Health Inc. were linked with 17 SEER registries data. HR + BC cases with lymph node negative (N0) or 1-3 positive LNs (N1) from 2004-2015 were analyzed. The Cochrane-Armitage trend test, logistic regression, Kaplan-Meier survival curve, and stratified Cox model were performed. Survival analysis was restricted to HR+/HER2- patients from 2010 to 2014, matched on propensity score.

RESULTS

ODX use increased substantially from 2004 to 2015 (N0: 2.0% to 42.7%; N1: 0.3% to 27.9%). Non-Hispanic black and Medicaid insured patients had lower odds of receiving ODX. N0 patients with moderately differentiated or 2.1-5.0 cm tumor and N1 patients with well-differentiated or < 2.0 cm tumor had higher odds of using ODX. The reported chemotherapy use decreased significantly with low and intermediate RS, and increased for high RS among N0 patients. ODX use was associated with better breast cancer-specific survival [hazard ratio (95% CI) N0 1.96 (1.60-2.41), N1 1.90 (1.42-2.54)] and overall survival [N0 2.06 (1.83-2.31), N1 1.72 (1.42-2.09)], especially in the first 36 months.

CONCLUSION

ODX use has increased significantly since 2004, nonetheless disparities remain, especially for racial/ethnic minorities and Medicaid insured patients. Administering chemotherapy based on ODX results has been improved among N0 patients. Patients receiving ODX had better survival than those not.

Workshop summary: Potential usefulness and feasibility of a US National Mesothelioma Registry

The burden and prognosis of malignant mesothelioma in the United States have remained largely unchanged for decades, with approximately 3200 new cases and 2400 deaths reported annually. To address care and research gaps contributing to poor outcomes, in March of 2019 the Mesothelioma Applied Research Foundation convened a workshop on the potential usefulness and feasibility of a national mesothelioma registry. The workshop included formal presentations by subject matter experts and a moderated group discussion. Workshop participants identified top priorities for a registry to be (a) connecting patients with high-quality care and clinical trials soon after diagnosis, and (b) making useful data and biospecimens available to researchers in a timely manner. Existing databases that capture mesothelioma cases are limited by factors such as delays in reporting, deidentification, and lack of exposure information critical to understanding as yet unrecognized causes of disease. National disease registries for amyotrophic lateral sclerosis (ALS) in the United States and for mesothelioma in other countries, provide examples of how a registry could be structured to meet the needs of patients and the scientific community. Small-scale pilot initiatives should be undertaken to validate methods for rapid case identification, develop procedures to facilitate patient access to guidelines-based standard care and investigational therapies, and explore approaches to data sharing with researchers. Ultimately, federal coordination and funding will be critical to the success of a National Mesothelioma Registry in improving mesothelioma outcomes and preventing future cases of this devastating disease.

Patient-Reported Outcomes Through 5 Years for Active Surveillance, Surgery, Brachytherapy, or External Beam Radiation With or Without Androgen Deprivation Therapy for Localized Prostate Cancer

Importance

Understanding adverse effects of contemporary treatment approaches for men with favorable-risk and unfavorable-risk localized prostate cancer could inform treatment selection.

Objective

To compare functional outcomes associated with prostate cancer treatments over 5 years after treatment.

Design, Setting, and Participants

Prospective, population-based cohort study of 1386 men with favorable-risk (clinical stage cT1 to cT2bN0M0, prostate-specific antigen [PSA] ≤20 ng/mL, and Grade Group 1-2) prostate cancer and 619 men with unfavorable-risk (clinical stage cT2cN0M0, PSA of 20-50 ng/mL, or Grade Group 3-5) prostate cancer diagnosed in 2011 through 2012, accrued from 5 Surveillance, Epidemiology and End Results Program sites and a US prostate cancer registry, with surveys through September 2017.

Exposures

Treatment with active surveillance (n = 363), nerve-sparing prostatectomy (n = 675), external beam radiation therapy (EBRT; n = 261), or low-dose-rate brachytherapy (n = 87) for men with favorable-risk disease and treatment with prostatectomy (n = 402) or EBRT with androgen deprivation therapy (n = 217) for men with unfavorable-risk disease.

Main Outcomes and Measures

Patient-reported function, based on the 26-item Expanded Prostate Index Composite (range, 0-100), 5 years after treatment. Regression models were adjusted for baseline function and patient and tumor characteristics. Minimum clinically important difference was 10 to 12 for sexual function, 6 to 9 for urinary incontinence, 5 to 7 for urinary irritative symptoms, and 4 to 6 for bowel and hormonal function.

Results

A total of 2005 men met inclusion criteria and completed the baseline and at least 1 postbaseline survey (median [interquartile range] age, 64 [59-70] years; 1529 of 1993 participants [77%] were non-Hispanic white). For men with favorable-risk prostate cancer, nerve-sparing prostatectomy was associated with worse urinary incontinence at 5 years (adjusted mean difference, -10.9 [95% CI, -14.2 to -7.6]) and sexual function at 3 years (adjusted mean difference, -15.2 [95% CI, -18.8 to -11.5]) compared with active surveillance. Low-dose-rate brachytherapy was associated with worse urinary irritative (adjusted mean difference, -7.0 [95% CI, -10.1 to -3.9]), sexual (adjusted mean difference, -10.1 [95% CI, -14.6 to -5.7]), and bowel (adjusted mean difference, -5.0 [95% CI, -7.6 to -2.4]) function at 1 year compared with active surveillance. EBRT was associated with urinary, sexual, and bowel function changes not clinically different from active surveillance at any time point through 5 years. For men with unfavorable-risk disease, EBRT with ADT was associated with lower hormonal function at 6 months (adjusted mean difference, -5.3 [95% CI, -8.2 to -2.4]) and bowel function at 1 year (adjusted mean difference, -4.1 [95% CI, -6.3 to -1.9]), but better sexual function at 5 years (adjusted mean difference, 12.5 [95% CI, 6.2-18.7]) and incontinence at each time point through 5 years (adjusted mean difference, 23.2 [95% CI, 17.7-28.7]), than prostatectomy.

Conclusions and Relevance

In this cohort of men with localized prostate cancer, most functional differences associated with contemporary management options attenuated by 5 years. However, men undergoing prostatectomy reported clinically meaningful worse incontinence through 5 years compared with all other options, and men undergoing prostatectomy for unfavorable-risk disease reported worse sexual function at 5 years compared with men who underwent EBRT with ADT.

Breast and colorectal cancer recurrence and progression captured by five U.S. population-based registries: Findings from National Program of Cancer Registries patient-centered outcome research

OBJECTIVES

Cancer recurrence is a meaningful patient outcome that is not captured in population-based cancer surveillance. This project supported National Program of Cancer Registries central cancer registries in five U.S. states to determine the disease course of all breast and colorectal cancer cases. The aims were to assess the feasibility of capturing disease-free (DF) status and subsequent cancer outcomes and to explore analytic approaches for future studies.

METHODS

Data were obtained on 11,769 breast and 6033 colorectal cancer cancers diagnosed in 2011. Registry-trained abstractors reviewed medical records from multiple sources for up to 60 months to determine documented DF status, recurrence, progression and residual disease. We described the occurrence of these patient-centered outcomes along with analytic considerations when determining time-to-event outcomes and recurrence-free survival.

RESULTS

Disease-free status was determined on all but 3.8 % of cancer cases. Among 14,458 cases that became DF, 6.1 % of breast and 13.0 % of colorectal cancer cases had a documented recurrence. Recurrence-free survival varied by stage; for stage II-III cancers at 48 months, 83.2 % of female breast and 69.2 % of colorectal cancer patients were alive without recurrence. The ability to distinguish between progression and residual disease among never disease-free patients limited our ability to examine progression as an outcome.

CONCLUSIONS

This study demonstrated that population-based registries given intense support and resources can capture recurrence and offer a generalizable picture of cancer outcomes. Further work on refining definitions, sampling strategies, and novel approaches to capture recurrence could advance the ability of a national cancer surveillance system to contribute to patient-centered outcomes research.

Trends in Incidence of Early-Onset Colorectal Cancer in the United States Among Those Approaching Screening Age

IMPORTANCE

Early-onset colorectal cancer incidence rates among patients aged 45 to 49 years have been considered much lower compared with the rates among patients aged 50 to 54 years, prompting debate about earlier screening benefits at 45 years. However, the observed incidence rates in the Surveillance, Epidemiology, and End Results (SEER) registries may underestimate colorectal cancer case burdens in those younger than 50 years compared with those older than 50 years because average-risk screening is generally not performed to detect preclinical cases of colorectal cancer. Finding steep incidence increases of invasive stage (beyond in situ) cases of colorectal cancer from age 49 to 50 years would be consistent with high rates of preexisting, undetected cancers in younger patients ultimately receiving a diagnosis of colorectal cancer after undergoing screening at 50 years.

OBJECTIVE

To assess the preclinical burden of colorectal cancer by analyzing its incidence in 1-year age increments, focusing on the transition between ages 49 and 50 years.

DESIGN, SETTING, AND PARTICIPANTS

Data from the SEER 18 registries, representing 28% of the US population, were used to conduct a cross-sectional study of colorectal cancer incidence rates from January 1, 2000, to December 31, 2015, in 1-year age increments (ages 30-60 years) stratified by US region (South, West, Northeast, and Midwest), sex, race, disease stage, and tumor location. Statistical analysis was conducted from November 1, 2018, to December 15, 2019.

MAIN OUTCOMES AND MEASURES

Incidence rates of colorectal cancer.

RESULTS

A total of 170 434 cases of colorectal cancer were analyzed among 165 160 patients (92 247 men [55.9%]; mean [SD] age, 51.6 [6.7] years). Steep increases in the incidence of colorectal cancer in the SEER 18 registries were found from 49 to 50 years of age (46.1% increase: 34.9 [95% CI, 34.1-35.8] to 51.0 [95% CI, 50.0-52.1] per 100 000 population). Steep rate increases from 49 to 50 years of age were also seen in all US regions, men and women, white and black populations, and in colon and rectal cancers. The rate ratio incidence increase in the SEER 18 registries from 49 to 50 years of age (1.46 [95% CI, 1.43-1.51]) was significantly higher than earlier 1-year age transitions. Steep rate increases in the SEER 18 registries were found from 49 to 50 years of age in localized-stage (75.9% increase: 11.2 [95% CI, 10.7-11.7] to 19.7 [95% CI, 19.0-20.3] per 100 000) and regional-stage (30.3% increase: 13.2 [95% CI, 12.7-13.8] to 17.2 [95% CI, 16.7-17.8] per 100 000) colorectal cancers. A total of 8799 of the 9474 cases (92.9%) of colorectal cancer in the SEER 18 registries from 2000 to 2015 that were diagnosed among individuals aged 50 years were invasive.

CONCLUSIONS AND RELEVANCE

Steep incidence increases between 49 and 50 years of age are consistent with previously undetected colorectal cancers diagnosed via screening uptake at 50 years. These cancers are not reflected in observed rates of colorectal cancer in the SEER registries among individuals younger than 50 years. Hence, using observed incidence rates from 45 to 49 years of age alone to assess potential outcomes of earlier screening may underestimate cancer prevention benefits.

Multiple Primary Cancers in the United States

BACKGROUND AND OBJECTIVES

The distribution of multiple primary cancers has been described previously using data from the Surveillance, Epidemiology, and End Results (SEER) Program. However, a complete picture regarding the distribution of multiple primary cancers in the United States is still lacking. The objective of the current study is to present a comprehensive description of multiple primary cancers in the United States.

MATERIALS AND METHODS

Invasive cancer cases (including in situ bladder cancers) diagnosed between 2001 and 2016 from 49 population-based state cancer registries in the United States were evaluated for this study. The sequence number central assigned to each tumor was used to determine whether a tumor was a first primary cancer or a subsequent multiple primary cancer. Tumors with a sequence number 00 or 01 were classified as the first primary cancer, while tumors with a sequence number 02 or above were classified as a multiple primary cancer. The percentage of multiple primary cancers was calculated by sex, age, race/ethnicity, cancer site, registry, and diagnosis year. In addition, the percentage of cancers diagnosed at a local stage among multiple primaries was compared with that among first primaries.

RESULTS

Overall, about 19.0% of cases were reported as multiple primary cancers; the percentage was higher among non-Hispanic Whites and among older patients. Bladder, melanoma of the skin, and lung cancers had the highest percentage of cases reported as multiple primaries. The percentage of multiple primary cancers also varied by registry and has been increasing over time. Cancers reported as multiple primaries were more likely to be diagnosed at a local stage than those reported as first primaries.

CONCLUSIONS

Cancers registered as multiple primaries are common in the United States, showing an increasing trend over time and wide variation by race/ethnicity, age, cancer type, and registry. The findings have some practical implications for cancer registries that collect data and for researchers conducting investigations using information on multiple primary cancers.

Association of Gastroesophageal Reflux With Malignancy of the Upper Aerodigestive Tract in Elderly Patients

Importance

Chronic inflammatory states have been linked to the development of malignancy. Gastroesophageal reflux disease (GERD) is a known risk factor for esophageal adenocarcinoma as the end result of chronic inflammatory changes.

Objective

To investigate the association of GERD with the risk of malignancy in the upper aerodigestive tract (UADT).

Design, Setting, and Participants

We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to conduct a case-control study of individuals in the United States who had been added from January 2003 through December 2011 and were 66 years or older. The study included patients diagnosed with malignancy of the larynx, hypopharynx, oropharynx, tonsil, nasopharynx, and paranasal sinuses. GERD was examined as an exposure. Controls were matched from a 5% random sample of Medicare beneficiaries without cancer. Multivariable unconditional logistic regression was performed.

Main Outcomes and Measures

Incidence of invasive malignancies of the UADT.

Results

A total of 13 805 patients (median [range] age, 74 [66-99] years; 3418 women [24.76%] and 10 387 men [75.24%]) with malignancy of the UADT were compared with 13 805 patients without disease and were matched for sex, age group, and year of diagnosis. GERD was associated with a greater odds of developing malignancy of the larynx (adjusted odds ratio [aOR], 2.86; 95% CI, 2.65-3.09), hypopharynx (aOR, 2.54; 95% CI 1.97-3.29), oropharynx (aOR, 2.47; 95% CI, 1.90-3.23), tonsil (aOR, 2.14; 95% CI, 1.82-2.53), nasopharynx (aOR, 2.04; 95% CI, 1.56-2.66), and paranasal sinuses (aOR, 1.40; 95% CI, 1.15-1.70).

Conclusions and Relevance

GERD is associated with the presence of malignancy of the UADT in the US elderly population. This epidemiological association requires further examination to determine causality and diagnostic utility.

Cancer survivor perspectives on sharing patient-generated health data with central cancer registries

PURPOSE

Central cancer registries collect data and provide population-level statistics that can be tracked over time; yet registries may not capture the full range of clinically relevant outcomes. Patient-generated health data (PGHD) include health/treatment history, biometrics, and patient-reported outcomes (PROs). Collection of PGHD would broaden registry outcomes to better inform research, policy, and care. However, this is dependent on the willingness of patients to share such data. This study examines cancer survivors' perspectives about sharing PGHD with central cancer registries.

METHODS

Three U.S. central registries sampled colorectal, non-Hodgkin lymphoma, and metastatic breast cancer survivors 1-4 years after diagnosis, recruiting them via mail to participate in one of seven focus groups (n = 52). Group discussions were recorded, transcribed, and thematically analyzed.

RESULTS

Most survivor-participants were unaware of the existence of registries. After having registries explained, all participants expressed their willingness to share PGHD with them if treated confidentially. Participants were willing to provide information on a variety of topics (e.g., medical history, medications, symptoms, financial difficulties, quality of life, biometrics, nutrition, exercise, and mental health), with a focus on long-term effects of cancer and its treatment. Participants' preferred mode for providing data varied. Participants were also interested in receiving information from registries.

CONCLUSIONS

Our results suggest that registry-based collection of PGHD is acceptable to most cancer survivors and could facilitate registry-based efforts to collect PGHD/PROs. Central cancer registry-based collection of PGHD/PROs, especially on long-term effects, could enhance registry support of cancer control efforts including research and population health management.

Abstract 4204: Racial difference in the risk of secondary bladder cancer following radiation therapy among prostate cancer patients

Purpose: Radiation therapy has been found associated with increased risk of secondary bladder cancer diagnosis among prostate cancer patients. It is unknown whether this association varies by race. This study aimed to investigate whether the association between radiation and secondary bladder cancer differs for white vs. black prostate cancer patients.

Methods: Data on white or black men who were diagnosed with primary localized prostate cancer in Louisiana between 2004 and 2013 and received radiation only or surgery only were obtained from Louisiana Tumor Registry. We excluded those who developed secondary bladder cancer within 180 days of the prostate cancer diagnosis or had follow-up less than 180 days or with unknown tumor size or Gleason score. Every patient was followed up until the end of 2016. The exposure variable was treatment (radiation only vs. surgery only). The outcome variable was the diagnosis of secondary bladder cancer and the time from prostate cancer diagnosis to secondary bladder cancer diagnosis or to loss to or end of follow-up. Covariates included age at diagnosis, marital status, insurance, census tract level poverty, tumor size, and Gleason score group. The competing risk survival analysis was applied. Death during follow-up was considered as competing event. Analysis was stratified by race.

Results: Of 10,861 white and 5,117 black prostate cancer patients, 52.7% and 56.8% received only radiation respectively. Regardless of race, patients who only received surgery were younger, more likely to be married, privately insured, live in low poverty census tract, and have smaller or lower grade tumor (P &lt;0.0001 for each in both racial groups). Among white prostate cancer patients who received surgery only vs. radiation only, 31 (0.60%) vs. 127 (2.22%) of them developed secondary bladder cancer (P &lt; 0.0001). Among black surgery only vs. radiation only patients, 14 (0.63%) vs. 29 (1.00%) of them developed a secondary bladder cancer (P = 0.16). The median follow-up time was 8.9, 8.0, 8.3, and 7.6 years for white surgery, white radiation, black surgery, and black radiation patients, respectively. After adjusting for covariates, the hazard ratio (95% confidence interval) [HR (95% CI)] of developing secondary bladder cancer was 2.70 (1.58, 4.61) for white radiation patients compared to white surgery patients; 0.95 (0.31, 2.89) for black radiation patients compared to black surgery patients. Compared to white surgery patients, the HR (95% CI) of developing secondary bladder cancer was 1.16 (0.61, 2.22) and 1.38 (0.74, 2.56) for black surgery and black radiation patients, respectively.

Conclusions: Radiation is associated with increased risk of secondary bladder cancer among white prostate cancer patients, but not among black patients. Early screening of secondary bladder cancer need to be emphasized among white prostate cancer patients who received radiation therapy.

Citation Format: Lu Zhang, Mei-Chin Hsieh, Qingzhao Yu, Xiao-Cheng Wu. Racial difference in the risk of secondary bladder cancer following radiation therapy among prostate cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4204.

Deep Transfer Learning Across Cancer Registries for Information Extraction from Pathology Reports

Automated text information extraction from cancer pathology reports is an active area of research to support national cancer surveillance. A well-known challenge is how to develop information extraction tools with robust performance across cancer registries. In this study we investigated whether transfer learning (TL) with a convolutional neural network (CNN) can facilitate cross-registry knowledge sharing. Specifically, we performed a series of experiments to determine whether a CNN trained with single-registry data is capable of transferring knowledge to another registry or whether developing a cross-registry knowledge database produces a more effective and generalizable model. Using data from two cancer registries and primary tumor site and topography as the information extraction task of interest, our study showed that TL results in 6.90% and 17.22% improvement of classification macro F-score over the baseline single-registry models. Detailed analysis illustrated that the observed improvement is evident in the low prevalence classes.

Early Versus Late Computed Tomography and Nasal Endoscopy in the Diagnosis of Nasopharyngeal and Paranasal Sinus Malignancy

Background

Chronic rhinosinusitis (CRS) and allergic rhinitis (AR) may be associated with an increased risk of subsequent diagnosis of nasopharyngeal carcinoma (NPC) or paranasal sinus cancer (PSC) in elderly Americans. The clinical utility of this association remains uncertain.

Objective

To compare early computed tomography (CT) or nasal endoscopy (NE) with late diagnostic studies for the diagnosis of NPC or PSC in elderly Americans with CRS or AR.

Methods

The Surveillance, Epidemiology, and End Results (SEER)-Medicare database was queried from 2003 to 2011 and included 150 088 Medicare beneficiaries. Patients with a diagnosis of CRS or AR were examined for either NE or CT performed within 6 months of the exposure diagnosis. The risk of a cancer diagnosis was determined between the early and the late diagnostic groups.

Results

The relative risk of early cancer diagnosis with NE was 1.98 (95% confidence interval [CI], 1.60–2.43). The number needed to detect (NND) a case of cancer with NE was 503 (95% CI, 387–718). The relative risk of an early cancer diagnosis using CT was 3.40 (95% CI, 2.85–4.06) and NND was 221 (95% CI, 194–255). The stage of NPC or PSC for the late diagnostic group was not different from those with early NE ( P = .458) or CT ( P = .497). Overall survival was not different between diagnostic groups for NE ( P = .789) or CT ( P = .425).

Conclusions

Early NE or CT is associated with a higher likelihood of cancer diagnosis in elderly individuals with a diagnosis of CRS or AR. The clinical utility of this association is limited due to the low prevalence of these malignancies and lack of difference in disease stage and overall survival between diagnostic groups.

Neighborhood Social Determinants of Triple Negative Breast Cancer

Triple Negative Breast Cancer (TNBC) is an aggressive, heterogeneous subtype of breast cancer, which is more frequently diagnosed in African American (AA) women than in European American (EA) women. The purpose of this study is to investigate the role of social determinants in racial disparities in TNBC. Data on Louisiana TNBC patients diagnosed in 2010–2012 were collected and geocoded to census tract of residence at diagnosis by the Louisiana Tumor Registry. Using multilevel statistical models, we analyzed the role of neighborhood concentrated disadvantage index (CDI), a robust measure of physical and social environment, in racial disparities in TNBC incidence, stage at diagnosis, and stage-specific survival for the study population. Controlling for age, we found that AA women had a 2.21 times the incidence of TNBC incidence compared to EA women. Interestingly, the incidence of TNBC was independent of neighborhood CDI and adjusting for neighborhood environment did not impact the observed racial disparity. AA women were more likely to be diagnosed at later stages and CDI was associated with more advanced stages of TNBC at diagnosis. CDI was also significantly associated with poorer stage-specific survival. Overall, our results suggest that neighborhood disadvantage contributes to racial disparities in stage at diagnosis and survival among TNBC patients, but not to disparities in incidence of the disease. Further research is needed to determine the mechanisms through which social determinants affect the promotion and progression of this disease and guide efforts to improve overall survival.

Detection of laryngeal carcinoma in the U.S. elderly population with gastroesophageal reflux disease

BACKGROUND

An association is suggested between gastroesophageal reflux disease (GERD) or laryngopharyngeal reflux (LPR) and malignancy of the larynx in elderly patients in the United States. Early detection with flexible fiberoptic laryngoscopy (FFL) or CT remains poorly defined.

METHODS

The Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database was queried from 2003 to 2011.A case-cohort design evaluated patients aged 66 and older with a diagnosis of GERD and/or LPR for the occurrence of FFL or CT within 6 months of the exposure diagnosis.

RESULTS

Of a total 156 426 Medicare beneficiaries, the relative risk of early cancer diagnosis with FFL was 14.61(95% confidence interval [CI], 13.59-15.70), corresponding to a number needed to detect (NND) a case of cancer of 13(95% CI, 13-14). The relative risk of an early cancer diagnosis with CT was 31.83 (95% CI, 29.57-34.26), with a NND of 5 (95% CI, 5-5).

CONCLUSIONS

Early FFL and CT are associated with a higher likelihood of laryngeal cancer diagnosis in elderly individuals with a diagnosis of reflux. Screening trials are necessary to establish this relationship.

Evaluating the Use of LAST 2-Tiered Nomenclature and Its Impact on Reporting Cervical Lesions in a Population-Based Cancer Registry

BACKGROUND

Since 2012, the Lower Anogenital Squamous Terminology (LAST) Project recommended a 2-tiered nomenclature, low-grade and high-grade squamous intraepithelial lesion (LSIL and HSIL), to replace the 3-tiered cervical intraepithelial neoplasia (CIN) system for HPV-associated lesions. Prior to 2019, preinvasive cervical lesions classified as CIN3, severe dysplasia, carcinoma in situ (CIS), and adenocarcinoma in situ (AIS) were considered reportable to the Louisiana Tumor Registry for a CIN3 project funded by the Centers for Disease Control and Prevention (CDC); but lesions classified exclusively as high-grade/HSIL based on the 2-tiered system were not considered reportable. Due to the terminology changes, we wanted to know whether pre-2019 reportable criteria need to be modified to capture all reportable precancerous cervical cases diagnosed in 2019 forward.

OBJECTIVES

To evaluate the utilization of LAST 2-tiered classification, low-grade and high-grade squamous intraepithelial lesion, and p16 immunohistochemistry (IHC) testing on cervical biopsy/surgical specimens, assess the search criteria needed to identify high-grade lesions for the CDC-funded CIN3 project, and assess the impact of underreporting cervical lesions caused by terminology changes.

METHODS

An equal number of abnormal/precancerous and normal cervical findings from biopsy pathology reports received in 2015 were randomly selected by an artificial intelligence (AI) search engine developed by Artificial Intelligence in Medicine Inc (AIM) using pre2019 search criteria. Selected pathology reports were reflagged for the reportability by AIM audit software based on 2019 search criteria and manually reviewed for the use of reportable terms including CIN3, severe dysplasia, CIS, AIS, highgrade/HSIL terminology, and CIN2 or CIN2-3 with positive p16 IHC testing. Cohen's kappa statistic was used to assess the agreement between AIM auto-coding and manual review. Positive predictive values (PPV) and sensitivity tests were computed to evaluate the reportable terms.

RESULTS

Six out of 9 surveyed laboratories used 2-tiered terminology on cervical biopsy pathology reports and 7 performed p16 IHC tests. Of 1,974 randomly selected reports from 5 laboratories, 987 were flagged as precancer by AI using pre-2019 search criteria. After adding the high-grade/HSIL term into pre-2019 search criteria, precancerous reports increased by 29%. After manual review, 41.6% of these cases were reportable precancerous cervical cases with a PPV of 0.65 (95% CI, 0.62-0.67) and 13.6% had p16 IHC performed.

CONCLUSIONS

Both the 2-tiered and 3-tiered nomenclature are needed to ensure complete identification of all reportable high-grade cervical lesions.