Surveillance, Epidemiology, And End Results Population Research Articles

Abstract PO3-29-05: Clinical risk score to predict overall survival and breast cancer recurrence in phase 3 TAILORx trial and the Surveillance, Epidemiology, and End Results (SEER) Database populations

Abstract Background: Overall survival (OS) is the gold standard and most relevant endpoint in oncology research and patient care. It underscores a need to develop methods to estimate the likelihood of mortality in breast cancer. Methods: We evaluated molecular and clinical measures that predicted mortality risk using multiple logistic regression and developed a clinical risk score (CS) in 8,905 patients from the phase 3 TAILORx trial. Association of CS with mortality in patients at low, intermediate, and high risks was analyzed by Kaplan-Meier and multivariable Cox proportional-hazards regression analyses. The score was validated in an independent population of 148,301 patients with node-negative, hormone-receptor (HR)-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer in the SEER database. Likelihood ratio (LR) test and c-index were used to compare prognostic information provided by CS and a 21-gene recurrence score (RS) to OS, invasive disease-free survival (DFS), recurrence-free interval (RFI), and distant recurrence-free interval (dRFI). Results: The 10-year mortality rates in low-risk, intermediate risk, and high-risk groups were 3.9% (95% CI 3.1 to 4.8), 37.2% (95% CI 23.4 to 51.1), and 75.8% (95% CI 54.1 to 97.4), respectively (P< 0.0001). Compared with low-risk score, intermediate (adjusted HR 45.5, 95% CI 32.2 to 64.9) and high-risk scores (HR 2026, 95% CI 1194 to 3483) were associated with risk of mortality, independent of demographics, tumor grade, RS, tumor size, PR, and treatment factors. The algorithm also predicted DFS, dRFI, and RFI outcomes (P< 0.0001 each), outperforming RS by both LR test and c-index estimate. Its validity was corroborated in the SEER population (intermediate risk: HR 9.3, 95% CI 6.7 to 13.3 and high-risk: HR 125.9, 95% CI 85.2 to 189.6). Conclusions: Clinical risk score robustly predicts risk of mortality in early-stage breast cancer populations. It provides more prognostic information than RS to survival and breast cancer recurrences. Citation Format: Sherry Yang, John Yu, Ninet Sinaii. Clinical risk score to predict overall survival and breast cancer recurrence in phase 3 TAILORx trial and the Surveillance, Epidemiology, and End Results (SEER) Database populations [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-29-05.

Effect of histological subtype on the efficacy of perioperative chemotherapy in pulmonary sarcomatoid carcinoma: a retrospective study based on SEER population.

The efficacy of perioperative chemotherapy (PC) in pulmonary sarcomatoid carcinoma (PSC) is controversial. We conducted this study to investigate the effect of different histological subtypes on the efficacy of PC in PSC patients. Clinicopathological data of 811 PSC patients of different histological subtypes were collected from the Surveillance, Epidemiology, and End Results (SEER) database. Kaplan-Meier method and log-rank test were used to evaluate the effects of PC on the overall survival (OS) and cancer-specific survival (CSS) in different subtypes of PSC patients. Propensity score matching (PSM) was used to reduce potential confounding effects. Subgroup analyses were conducted to further investigate the efficacy of PC in patients with different characteristics. A total of 210 (25.89%) enrolled PSC patients received PC. PC was not associated with OS or CSS benefit in pleomorphic carcinoma, giant cell carcinoma, or spindle cell carcinoma patients, neither before nor after matching. But survival benefit of PC was observed in carcinosarcoma patients both before (5-year OS: 48.79% vs. 38.75%, P=0.01) and after (5-year OS: 51.29% vs. 17.54%, P=0.003) matching. Subgroup analyses showed that in patients whose tumor larger than 4 cm, PC was still associated with improved survival in carcinosarcoma, but not in the other histological subtypes of PSC. The efficacy of PC varies between different subtypes of PSC. Survival benefit of PC was only observed in carcinosarcoma patients, but not in pleomorphic carcinoma, giant cell carcinoma, or spindle cell carcinoma patients. Histological subtype should be considered when treating PSC patients with PC.

Integrating the 31-gene expression profile test into clinical decision-making to guide risk-aligned care decisions for patients with stage I-III cutaneous melanoma: NCI-SEER Analysis.

6601 Background: Cutaneous melanoma (CM) management depends on an estimate of recurrence risk or death. Because of recognized disease risk and drug efficacy, pembrolizumab immunotherapy has been approved for stage IIB-C patients. However, it is a treatment with potential financial and biological toxicities with 16% of patients experiencing severe treatment related adverse events. Improved risk stratification could identify individuals for whom more selective management would be appropriate. In collaboration with the NCI Surveillance, Epidemiology, and End Results (SEER) program, which includes information on patients tested with a 31-gene expression profile (31-GEP) test in an unbiased and unselected manner, we confirm the 31-GEP adds independent prognostic value in patients with stage I-III CM, and investigated whether the 31-GEP could identify stage IIB-C patients who might reasonably avoid adjuvant therapy while also identifying patients with stage I disease who might benefit from more aggressive diagnostic and/or therapeutic intervention. Methods: SEER registries linked CM cases diagnosed 2013-2018 to 31-GEP test results provided by Castle Biosciences (9,207 matched) using a registry trusted third party (honest broker), including subsets of stage IIB-IIC (n=775) and stage I (n=5,651) melanoma. The primary endpoint was melanoma-specific survival (MSS). The 31-GEP risk-groups included Class 1A (low), Classes 1B/2A (intermediate), and Class 2B (high). Each was compared using Kaplan-Meier analysis and the log-rank test. Results: In the overall cohort, patients with a Class 1A result had a significantly higher 5-year MSS than a Class 1B/2A or Class 2B result (97.4% vs. 93.5% vs. 83.3%, p<0.001). In multivariable analysis, a Class 2B result (HR=3.26, p<0.001) and positive lymph node status (HR=3.41, p<0.001) were the strongest predictors of diminished MSS. In stage IIB-C, 18.8% (146/775) of patients received a Class 1A result with an associated 5-year MSS of 89.1% (vs. Class 2B: 76.6%, p=0.001), despite the absence of pembrolizumab approval for the evaluated SEER population. Alternatively, 4.6% (260/5,651) of patients with stage I disease had a Class 2B result, with an associated 5-year MSS of 92.3% (vs. Class 1A: 98.0%, p<0.001). Conclusions: The SEER database allowed a non-biased analysis of a large cohort of patients with stage I-III CM who were tested with the 31-GEP. The 31-GEP identifies patients, including those with stage IIB-C disease, with favorable outcomes and could help physicians and patients decide whether to pursue adjuvant therapy. Moreover, the 31-GEP identities high-risk patients, including those with stage I disease, for whom more aggressive follow-up and early intervention should be considered.

Abstract P4-03-13: AN EVALUATION OF BREAST CANCER RECURRENCE DATA REPORTED TO THE NCI SEER PROGRAM

Abstract Background: There are currently 18.1 million cancer survivors in the U.S. and the number is expected to increase to 22.5 million in the next ten years. The increasing number of survivors also means that the population at risk for cancer recurrence will increase. Recurrence significantly impacts patients’ health outcomes because there are limited treatment options. Currently, there are limited population level recurrence data in cancer registries world-wide. Most recurrence estimates have been derived from randomized clinical trials (RCTs) where the patient population only represents 4% of all cancer patients. Since 2018 the National Cancer Institute’s Surveillance, Epidemiology and End Results (SEER) program has been collecting abstract-based recurrence information. Our goal is to use the SEER recurrence data sources to develop a high-quality population-based recurrence database to disseminate to the public for future research. Measuring recurrence from population-based cohorts can enhance our understanding of what clinical, demographic, and geographic factors impact the risk of recurrence and can inform researchers on developing prevention strategies. Methods: In November 2021 we obtained a dataset that included recurrence information such as type of first recurrence (defined by the Standard for Oncology Registry Entry manual) and date of first recurrence. To evaluate the burden of breast cancer recurrence in the U.S. SEER population, we described the cohort of women diagnosed with breast cancer from 2010 to 2019. We then evaluated the patients with information on the status of recurrence including those that never experience recurrence and those that were never disease free (NDF). We measured frequencies of recurrence by multiple clinical variables such as age at diagnosis, race, stage at diagnosis, and tumor subtype. We followed patients from their date of diagnosis to the event defined by first recurrence, second primary, death, or last contact. We used the Kaplan-Meier method to estimate DFS. Results: The analytic dataset included 412,920 female breast cancer cases diagnosed between 2010 and 2019. Most women diagnosed with breast cancer were between 40 and 64 years old (53%, n=218,760), were White (78%, n= 320,555), were diagnosed with localized tumors (52%, n=213,824), were HER2 negative (HER2-) hormone receptor positive (HR+) (56%, n=230,874), and received treatment (95%, n=390,959). During the observation period, 67% (n = 278,098) of women did not have a recurrence, 3% (n=10,922) had a recurrence, and 16% (n=66,319) were NDF. After excluding patients with NDF status, we observed that the age group with the highest frequency of recurrence was 20-39 (9%). For reference, 4% of women in the 40 to 64 age group has a recurrence event during the observation period. Black women had higher recurrence (6%) compared to white women (3%). Triple negative breast cancers had the highest recurrence (9%) out of all subtypes. When we measured DFS, we observed significant differences by race, independent of tumor subtype. For example, in patients with HER2+/HR+ tumors, the percentage of patients at 120 months who did not experience a DFS event were 73.6% (95% CI, 72.3% to 74.9%) in white women and 67.5% (95% CI, 63.9% to 70.8%) in black women. Discussion & Conclusion: We are reporting population-based recurrence estimates for breast cancer obtained through the SEER program. This is the largest breast recurrence database in North America. The size of the database allowed precise estimation of recurrence stratified by demographics, subtype, and extent of disease. We observe trends consistent with previous reports and have identified some novel findings that warrant further study. Our long-term goal is to assess how DFS estimates derived from a population-based cohort compares to previously published estimates which are largely derived from RCTs. Citation Format: Esmeralda Ramirez-Pena, Serban Negoita. AN EVALUATION OF BREAST CANCER RECURRENCE DATA REPORTED TO THE NCI SEER PROGRAM [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-03-13.

Age is a major determinant for poor prognosis in patients with pilocytic astrocytoma: a SEER population study.

Background Pilocytic astrocytomas (PAs) are central nervous system tumors with variable prognosis and poorly understood risk factors. Little evidence exists regarding the effect of age on mortality in PA. Therefore, we conducted a thorough characterization of PA in the US. Methods We queried the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2018 to extract age-adjusted incidence rate (AAIR), age-adjusted mortality rate (AAMR), and survival data on PA. The age group comparisons for each measure varied depending on available SEER data. We compared trends in AAIR and AAMR by two age groups (children, 0-19years; adults, 20 + years) and by sex. The cumulative incidence function and the Fine-Gray competing risk model were applied by 0-19, 20-39, 40-59, and 60 + years of age groups. Results This study included 5211 incident PA and 462 PA-specific deaths between 2000 and 2018. Trends in AAIRs and AAMRs were almost constant between 2000 and 2018. Average AAIRs had a sharp peak in 1-4years of age groups, whereas AAMRs had a gradual peak in 80-84years of age groups. Age groups, tumor location, and race/ethnicity were significantly associated with PA-specific death, whereas only age was associated with other cause of deaths. Conclusions Trends in AAIRs and AAMRs were constant regardless of age. PAs in older populations, especially over 60years old, have higher incidence of death than those in younger populations.

Development and validation of a dynamic survival nomogram for metastatic non-small cell lung cancer based on the SEER database and an external validation cohort.

BackgroundLimited efficacy and poor prognosis are common in patients with metastatic non-small cell lung cancer (NSCLC). An accurate and useful nomogram helps the clinician predict the prognosis of the patients. However, there has been no previous report on the nomogram specially for predicting the overall survival (OS) of metastatic NSCLC patients.MethodsA total of 18,343 patients diagnosed with metastatic NSCLC in the Surveillance, Epidemiology, and End Results (SEER) database were included and divided into the training cohort (n=12,840) and the internal validation cohort (n=5,503), and 242 patients in Renji Hospital were additionally enrolled as the external validation cohort. Demographical, clinical, and OS data were collected. A Cox proportional hazards regression model was used to develop a nomogram based on the training cohort. To validate the nomogram, we applied C-indexes, calibration curves, receiver operating characteristic (ROC) curve, decision curve analysis (DCA), and a Kaplan-Meier survival curve.ResultsThe multivariate Cox regression model found that there were a total of 16 independent risk factors for OS of the patients (all 16 factors showed P<0.001), which were integrated into the nomogram with a C-index of 0.702 [95% confidence interval (CI): 0.684–0.720]. The nomogram also exhibited good prognostic value in the internal validation cohort (C-index =0.699, 95% CI: 0.673–0.725) and external validation cohort (C-index =0.695, 95% CI: 0.653–0.737). The ROC and Kaplan-Meier survival curve analyses demonstrated a high discriminative ability. High-risk patients had significantly less favorable OS than low-risk patients in the SEER population and external validation cohort (both P<0.001). The DCA analysis showed that the nomogram provided better prognosis prediction than the tumor-node-metastasis (TNM) staging system.ConclusionsWe constructed and validated a dynamic nomogram with 16 variables based on a large-scale population of SEER database to predict the prognosis of metastatic NSCLC patients. The nomogram is expected to provide higher predictive ability and accuracy than the TNM staging system.

154 Increasing incidence of cutaneous t-cell lymphoma in the United States: A SEER population data analysis

Recent incidence patterns of cutaneous T-cell lymphoma (CTCL) in the United States are not well described. Our goal was to evaluate incidence trends of CTCL by subtype and examine associations to sex, age, race/ethnicity, and socioeconomic status (SES). Using data from the Surveillance, Epidemiology, and End Results (SEER) Registry, we calculated age-adjusted incidence rates and annual percentage change (APC) in incidence of CTCL. We identified 14,942 patients with a new diagnosis of CTCL between 2000 and 2018.

Survival Outcomes Related to Treatment Modalities in Patients With Oropharyngeal Squamous Cell Carcinoma.

More patients with oropharyngeal squamous cell carcinoma (OPSCC) in Eastern countries receive surgically inclusive treatment (SIT), while most patients in Western countries receive nonsurgical treatment (NST). The optimal treatment modality for OPSCC patients remains controversial. A total of 153 consecutive OPSCC cases diagnosed between 2009 and 2019 in West China Hospital (WCH) and 15,400 OPSCC cases from the Surveillance, Epidemiology, and End Results (SEER) database (2000-2017) were obtained. Clinical characteristics, treatments, and survival outcomes were retrospectively collected. We constructed Kaplan-Meier curves and performed univariate (UVA) and multivariate (MVA) analyses to compare the prognosis of OPSCC patients among the WCH, SEER Asian, and SEER all ethnic populations by different treatment modality, human papilloma virus (HPV) infection status, age, and tumor stage. Overall, the proportions of patients with younger age, advanced tumors and HPV-negative status, and receiving SIT in WCH population were higher than those in the SEER all ethnic population, while the proportions in the SEER Asian population were between those of the other two populations. We observed consistent beneficial effects of SIT on the overall survival (OS) in OPSCC patients in all three populations (SEER Asian: MVA, hazard ratio (HR): 0.2, p < .001; SEER all ethnic: MVA, HR: 0.46, p < .001; WCH: UVA, HR: 0.62, p = .071), and HPV-negative Asian patients showed greater benefits from the SIT than HPV-positive Asian patients (HPV Negative: HR: 0.16, p = .005; HPV positive: HR = 0.28, p = .059). Male was a risk factor for reduced OS in OPSCC patients in the WCH population (HR: 3.17, p = .043), but was a protective factor in the SEER population (HR: 0.8, p = .002), which might be related to the differences of HPV infection status. Even though differences in patient characteristics existed between the Chinese, American, and Asian American populations, our ten-year real-world data and SEER data suggested that patients with OPSCC who received SIT had a better prognosis than those who received NST.

Trends in the cause of death among patients with bladder cancer in the US SEER population, 1992-2018.

Bladder cancer (BC) survival has shown no significant improvement. This study investigated the trends in the common causes of death among patients with BC to improve the management and survival of BC. The Surveillance, Epidemiology, and End Results (SEER) (1992-2018) database was utilized to get the data of BC patients. We presented the proportion of six common causes of death in BC patients. We calculated the annual incidence of death due to the six most common causes and analyzed temporal trends in mortality rates using joinpoint regression. The competitive risk model was utilized to analyze the risk factors for death of BC and other causes. 198037 BC patients were enrolled. BC was the most common cause of death (30.62%), followed by other cancers (22.22%), circulatory diseases (20.28%), non-disease causes (11.58%), other non-cancer diseases (8.29%), and respiratory diseases (7.01%). However, the proportion of cases dying from BC gradually decreased from 44.87% in 1992-1996 to 26.74% in 2012-2018. The proportion of deaths due to BC decreased gradually with survival time from diagnosis. Age-standardized temporal trends present an initial increase in BC-specific and other-cause mortality rates. Advanced stage and older age were the most influential risk factors for BC-specific and other-cause death, respectively. Although BC was still the leading cause of death, other causes, especially other cancers and circulatory diseases, gradually became more critical. The management of other comorbid conditions will be a crucial part of the treatment for BC patients, especially for those with prolonged survival and NMIBC tumors.

Studying the association between breast cancer and renal cell carcinoma.

There are case reports of patients with both primary breast cancer (BC) and renal cell carcinoma (RCC). We explore the association between these two malignancies using SEER population data and our institutional records. We studied the association between BC and RCC in the 2000-2016 Surveillance, Epidemiology, and End Results (SEER) database. We then reviewed our hospital records of patients with both BC and RCC and collected information including personal and family history of cancers, genetic testing, and patient outcomes. Of the 813,477 females diagnosed with BC in the SEER database, 1914 later developed RCC. The risk of developing RCC was significantly increased within the first 6months, 7-12months, and 1-5years following BC diagnosis with standardized incidence ratios (SIRs) of 5.08 (95% CI 4.62-5.57), 2.09 (95% CI 1.8-2.42), and 1.15 (95% CI 1.06-1.24), respectively. Of 56,200 females with RCC, 1087 later developed BC. The risk of developing BC following RCC was elevated within the first 6months (SIR of 1.45 [95% CI 1.20-1.73]). For our hospital patients, 437 had both BC and RCC. 427 (97.71%) were female, and 358 (81.92%) were white, and breast cancer was diagnosed before RCC in 246 (56.3%) patients. There were 15 germline mutations in those with genetic testing. Our findings suggest that BC patients are at higher risk of developing RCC and vice versa. BC tended to precede RCC, and patients frequently had personal histories of other malignancies and a family history of cancer, particularly, BC.

Sex disparities in thyroid cancer: a SEER population study.

The incidence and mortality of thyroid cancer vary based on race as well as gender. Both gender thyroid cancer patients give variable clinical characteristics, such as tumor size and distant metastasis. However, sex differences in the prognosis of thyroid cancer remain controversial. Therefore, the present study explored the relationship between gender and prognosis of patients with thyroid cancer for conducive personalized treatment. A retrospective analysis was carried out on patients with pathologically proven thyroid cancer from the Surveillance, Epidemiology, and End Results (SEER) database. The gender disparities in the prognosis of different cohorts, derived by propensity score matching were investigated using Cox proportional hazards models and Kaplan-Meier curves. Among the studied 41,270 female and 13,188 males with thyroid cancer, gender was an independent prognostic factor for overall (OS) and cancer-specific (CSS) survival (HR =1.632, 95% CI: 1.499-1.777, P<0.001; HR =1.473, 95% CI: 1.245-1.741, P<0.001). Though, male patients had a larger tumor size (17.4 vs. 23.5 cm) and a larger proportion of metastasis [lymph nodes (LNs): 33.2% vs. 21.0%; distant: 2.3% vs. 0.9%], female had a higher incidence and earlier age diagnosis with thyroid cancer (48.0 vs. 52.5 years old). Survival Time (in months) of male patients was also significantly lower than female patients (72.4 vs. 76.8 months). In the Kaplan-Meier curves of cohorts derived by propensity score matching, OS and CSS declined much sharply for male (P<0.001). The mean number (2.0 vs. 4.0) and mean ratio (0.192 vs. 0.297) of positive nodes supported worse prognosis for male patients. Whereas factors including race, age, surgery, histology recodes, T, N, M stage and combined summary stage affected the CSS of male and female patients, however plus median income had an extra impact on male population (≥$55,000 vs. <$55,000: HR =0.739, 95% CI: 0.574-0.953, P=0.020). Our study demonstrated that male patients had a prognostic factor for poorer OS and CSS. Other factors including race, age, income, histological type, surgery, T, N, M stage influenced OS of male and female thyroid cancer patients. Interestingly, race had no impact on CSS of thyroid cancer patients, whereas median income affected only the male patients CSS.

Normal Life Expectancy for Polycythemia Vera Patients Is Possible

▪Introduction: Polycythemia vera (PV), a chronic myeloproliferative neoplasm, is characterized by shortened overall survival (OS) due to potentially fatal thrombosis, progression to myelofibrosis (MF), and acute leukemia. Treatment with aspirin, attentive phlebotomy (PHL) and cytoreductive agents can prevent thrombosis; but it is unknown whether available treatments can achieve a normal life-span for PV patients (pts). Our objective was to assess the extent by which PV alters OS of pts receiving available treatment at our center and in the community.Methods: We obtained demographics and survival data of adult PV pts from our Weill Cornell Medicine (WCM) Research Database Repository (PV-WCM) diagnosed from 1974-2020 as previously described (Abu-Zeinah et al. Leukemia 2021); and from the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) Program, 2001-2017. PV-WCM thrombosis and MF progression data were available. WCM institutional review board approval and SEER data use agreement were in place. OS of PV pts was estimated using Kaplan-Meier (KM) analysis. Propensity Score Matching was used to identify a subgroup of SEER pts (PV-SEER) with similar age, sex, and race demographics to PV-WCM for comparison of OS. Cox proportional hazards was used for multivariable analysis (MVA) of survival. Actuarial survival of the US population matched by age, sex, race, and year-of-birth was obtained from US Census. Observed OS was compared to actuarial OS using one-sided log-rank test.Results: We identified 470 PV-WCM and 16,492 PV pts from SEER. The median OS of the PV-WCM cohort and the SEER population was 26.6 and 12.8 years, respectively, but the groups differed by age and sex (Fig 1A). PV-WCM and matched PV-SEER, however, were well balanced by age, sex, and race (standardized mean difference <0.1). The median OS of PV-WCM was 10.8 years longer than PV-SEER (p<0.001, Fig 1B). MVA confirmed 65% lower mortality in the PV-WCM group relative to the PV-SEER population (p<0.001, Fig 1C). PV-SEER OS was shorter than actuarial OS of the US population (mortality HR 2.47, p<0.001), whereas PV-WCM OS was not significantly different than expected (mortality HR 1.15, p=0.136). Excess late mortality was observed in PV-WCM after 17 years, potentially due to increased incidence of MF progression, affecting almost 50% of PV pts after 25 years (Fig 1D).Discussion: These data point to the success of attentive management with aspirin, targeted PHL, and cytoreductive therapy at a specialty center in reducing the mortality of PV, owing in part, to ELN and NCCN treatment recommendations for preventing fatal thrombosis. Although excess thrombosis was not eliminated in PV-WCM pts, the incidence rate was low after the first two years (Fig 1D). In contrast, the incidence of MF progression increased, potentially contributing to excess late mortality. Current PV risk stratification and treatment recommendations neither predict nor aim to mitigate progression to MF, and do not satisfy the needs of younger, low-risk pts who are often symptomatic and harbor a lifetime threat of progression (Fig 1E). The recent Low-PV study showed that low-risk pts may also benefit from cytoreductive therapy with an agent such as interferon-alpha (rIFNα) over PHL alone (Barbui et al. Lancet Hematol 2021). PV-WCM pts treated with rIFNα (n=137), previously shown to have improved MF-free survival and OS (Abu-Zeinah et al. Leukemia 2021), had an OS similar to the matched US population (p=0.33, Fig 1F); whereas pts not treated with rIFNα had modestly shortened OS (HR 1.26, p=0.03). Because PV progression is among the leading causes of late morbidity and mortality, and available therapy can mitigate this, it may be time to reconsider and improve upon PV risk stratification and treatment.Conclusion: This large population-based and single center study shows that although PV survival remains shortened, normal life expectancy for PV pts is possible with available care. The discrepancy between survival in the community and at a specialized center highlights the need for greater community outreach, health equity, and education regarding treatment standards for PV. Hopefully, this study motivates development of new PV risk stratification and treatment recommendations that focus not only on hematologic control and thrombosis prevention, but also on preventing MF progression, improving OS, and restoring normal life expectancy. [Display omitted] DisclosuresAbu-Zeinah: PharmaEssentia: Membership on an entity's Board of Directors or advisory committees. Silver: Abbvie: Consultancy; PharamEssentia: Consultancy, Speakers Bureau. Scandura: CR&T (Foudation): Research Funding; European Leukemia net: Honoraria, Other: travel fees ; MPN-RF (Foundation): Research Funding; Constellation: Research Funding; Abbvie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. OffLabel Disclosure:Interferon-alpha is used off label in the treatment of polycythemia vera.

Differences in lymphoma patients between chimeric antigen receptor T-cell therapy trials and the general population.

Chimeric antigen receptor (CAR)-T cell therapies appear to be promising treatments for non-Hodgkin's and B-cell lymphoma. However, several CAR-T therapies approved by the US Food and Drug Administration have only been tested for efficacy in relatively few single-arm clinical trials with small sample sizes. We sought to examine the differences between patients in these trials and the general population of patients with non-Hodgkin's and B-cell lymphoma. Five hundred and twenty-two patients from 15 CAR-T trials found in a systematic review and 417,492 patients from the Surveillance, Epidemiology, and End Results (SEER) Program database were compared. CAR-T study participants appeared to be younger (46.7% under 70 years old vs. 42.2%), more male (68.0% vs. 55.7%), and followed for a shorter period of time compared to patients in the SEER population (mean [M] 45.6 months, 95% confidence interval [CI] 17.7 to 63.3 months follow-up vs. M 57.1 months, 95% CI 57.0 to 57.3 months survival). CAR-T study participants may differ significantly from the general population of patients with non-Hodgkin's and B-cell lymphoma. Effectiveness of CAR-T therapies in the general population of lymphoma patients may differ from effectiveness demonstrated in trials. Newly created CAR-T patient registries are essential to establishing population-level effectiveness of the therapies.

Abstract PS7-29: Racial disparities in breast cancer outcomes: A SEER population based study

Abstract Introduction: Breast cancer (BC) is the second leading cause of female cancer-related mortality in the United States. Despite advances in therapies which have significantly lowered BC mortality rates over the years, the decline in mortality in African American/Black women continues to lag behind. Black women with BC continue to do poorly compared to non-Hispanic White women. The etiologies underlying this disparity are multifactorial and still remain unclear. Our study aims to investigate the differences in clinical and socioeconomic characteristics (cts) between Blacks and Whites and evaluate their prognostic value on long-term outcomes. Methods: We conducted a retrospective, population-based analysis utilizing the Surveillance, Epidemiology, and End Results (SEER) database, and studied BC pts from 1975-2017. Different demographic and clinical cts were analyzed by race. Univariate (UV) and multivariable (MV) analyses were performed to evaluate the associations of race with disease-specific survival (DSS), overall survival (OS) using Cox regression models. Results: A total of 816,763 pts were analyzed (n=720,144 White, n=966,19 Black). As compared to White BC patients (pts), Black pts were younger in age (median age 57 years vs. 62 years), with a higher tumor stage (III/IV; 25% vs. 16.3%) and a higher tumor grade (III/IV; 52.2% vs. 35.3%) at disease presentation (p&amp;lt;0.001). Furthermore, Black pts were more likely to present with triple-negative (23.9% vs. 11.1%) and Her-2+ BC (20.6% vs. 16.3%) than White pts (p&amp;lt;0.001). Black pts were also more likely to be uninsured (3.4% vs.1.1%), single (64.3% vs. 42.5%), and less likely to receive breast surgery (90.7% vs. 95.4%) as compared to White pts (p&amp;lt;0.001). A higher proportion of Black pts lived in urban areas (93.9% vs. 87.9%) and had a higher county-level availability of hospitals with oncology services per million population (median 126 vs. 120.6) than White pts (p&amp;lt;0.001). Black pts had a worse DSS (Hazard Ratio (HR) 1.68, 95% CI 1.66-1.71)) and OS (HR 1.30, 95% CI 1.29-1.31) compared to White pts (p&amp;lt;0.001). The racial differences in DSS and OS remained significant on both UV and MV analyses while controlling for each demographic and clinical variable. Conclusions: Our study confirms that Black pts with BC have a worse OS and DSS as compared to White pts. These disparities could be partially explained by observed differences in underlying aggressive tumor biology (stage, grade, hormone receptor status, age at diagnosis), lower access to health care (insurance status and % receiving breast surgery) and weaker social support systems (% single). However, the fact that the survival disparity persisted even after controlling for these factors suggests that there are potentially more variables driving these racial differences in breast cancer outcomes. We note that more Black pts resided in UA and had access to hospitals with oncology services reflecting that there could be disparities in health care utilization patterns. Further studies investigating these variables would be of paramount importance in order to identify underlying causes for poorer outcomes in Black pts and help eliminate this survival gap. Citation Format: Medhavi Gupta, Rohit Gosain, Maithreyi Sarma, Stuthi Perimbeti, Kristopher Attwood, Wenyan Ji, Shipra Gandhi, Yara Abdou. Racial disparities in breast cancer outcomes: A SEER population based study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS7-29.

Development and validation of an individualized nomogram for predicting pancreatic adenocarcinoma-specific survival: a SEER population analysis of 5,805 patients.

A model to predict the overall survival (OS) of pancreatic adenocarcinoma (PAC) is required in consideration of its inferior prognosis. The patients diagnosed with pancreatic cancer between 1975 and 2016 in the Surveillance, Epidemiology, and End Results (SEER) database was used as raw data. A training cohort and a verification cohort were used for internal validation and external validation, respectively. The nomogram model was constructed to predict the OS. A total of 5,805 patients with PAC from 2010-2015 were analyzed. Most patients were over 65 years old (61.8%), white (81.2%), in stage IIA, IIB (49.0%), and IV (32.4%), less than 50 mm in diameter (80.2%). PAC patients with wide involvement range, no metastasis, and infiltration range more than 300 accounted for 58.2%, 67.6%, 78.2%, respectively. The vast majority of the PAC patients (90.9%) did not receive primary site surgery. Most of the PAC patients (68.1%) received chemotherapy and only 25.8% of PAC patients received radiotherapy. The overall mean survival time, overall median survival time and overall survival rate were 15.1 months, 10.0 months, and 16.7%, respectively. Our nomogram that based on age, chemotherapy, grade, Radiation sequence with surgery, Radiation recode, RX Summ-Surg Prim at Site (surgery that removes and/or destroys primary tumor performed as part of the first course of therapy), size, and stage was of well prediction ability.

Abstract C087: Differences in estrogen receptor-negative breast cancer by race and SEER registry

Abstract Background: Estrogen receptor-negative (ERN) breast cancer is an early-stage, more aggressive breast cancer subtype. ERN breast cancer incidence is highest in African American women compared with other races, as well as in southern US regions. However, recent studies have shown that ERN breast cancer incidence is declining for women overall. We sought to understand whether ERN breast cancer is declining similarly in all races, age groups, and SEER registries. Methods: Data from Surveillance, Epidemiology, and End-Results (SEER) Program 13 registries from 1990-2014, as well as the remaining five SEER 18 registries from 2000-2014, were used to observe ERN breast cancer incidence rates among non-Hispanic white (NHW) and non-Hispanic black (NHB) women by age group (30-39, 40-49, 50-69, 70-84 years). Age-period-cohort modeling was used and extended, allowing for regional heterogeneity by population (i.e. SEER registry), to estimate differences in longitudinal age trends and net drifts by race and SEER registry. Results: Among all age groups within all SEER 18 registries, ERN breast cancer incidence rates were higher for NHB compared to NHW women. Furthermore, ERN rates have been decreasing for both NHW and NHB women. For the entire SEER population, women ages 40-49 years compared to other age groups demonstrated the fastest declines in ERN breast cancer incidence, with a net drift of -3.5%/year (95% CI: -4.0, -3.1) for NHW women, and -3.1%/year (-3.8, -2.2) for NHB women. Among the youngest women (30-39 years), ERN rates have declined faster in NHB compared to NHW women, with net drifts of -2.7%/year (-3.5, -1.8) and -1.7%/year (-2.2, -1.2), respectively; for all other age groups, rates have declined faster for NHW compared to NHB women. For NHW women of all age groups, there was relatively little between-registry variability in trends, with net drift standard deviations of 0.5%/year (0.02, 1.3) for 30-39 years, 0.8%/year (0.4, 1.2) for 40-49 years, 0.6%/year (0.4, 0.9) for 50-69 years, and 0.7%/year (0.4, 1.1) for 70-84 years. For NHB women, rates have also decreased similarly across registries in the youngest age group (30-39 years), with between-registry net drift standard deviation of 0.4%/year (0.01, 1.5); however, for NHB of all other age groups, there was more between-registry variability with net drift standard deviations of 1.0%/year (0.1, 2.1) for 40-49 years, 1.0%/year (0.4, 1.8) for 50-59 years, and 1.2%/year (0.4, 2.4) for 70-84 years. These standard deviations reflect that ERN rates have been declining more slowly in southern registries (i.e., Greater Georgia, Atlanta, Rural Georgia, and Louisiana) for older NHB women. Conclusion: Among NHW women, decreases in ERN breast cancer incidence have been similar across age groups and SEER registries. Among NHB women, ERN rates have also been decreasing; however, these decreases vary by age and SEER registry, which may offer etiologic clues. Further investigation is needed to understand which factors are contributing to these trends. Citation Format: Brittny C. Davis Lynn, Pavel Chernyavskiy, William F. Anderson, Gretchen L. Gierach, Philip S. Rosenberg. Differences in estrogen receptor-negative breast cancer by race and SEER registry [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 C087.

Generalizability of national survival estimates to adolescent and young adults with cancer at an urban safety-net institution.

140 Background: Survival estimates for adolescent and young adults (AYAs) with cancer in the US are largely derived from the Surveillance, Epidemiology, and End Results (SEER) program, which does not include certain geographic areas with diverse populations such as Texas. The generalizability of these estimates to other populations is unknown but may be useful for identifying high-risk populations. Therefore, we aimed to assess the generalizability of survival estimates for AYAs with cancer in SEER to AYAs with cancer at an urban safety-net institution in Texas. Methods: We used data from the SEER program and the JPS Oncology and Infusion Center, an accredited Comprehensive Community Cancer Program. Eligible individuals in both populations were aged 18 – 39 years and diagnosed with a first primary cancer between 2008 and 2015. We computed entropy-balanced weights to adjust the baseline distribution of age, gender, race/ethnicity, insurance status, marital status, cancer type, and stage of the SEER population to our safety-net population. We used these weights to compute Kaplan-Meier estimates of expected survival for our safety-net population. We subsequently estimated the absolute survival difference between expected and observed (i.e. actual) survival and corresponding 95% confidence limits (CL). Results: Our safety-net population comprised 833 AYAs with cancer, of whom 60% were female, 62% were racial/ethnic minorities, and 66% were uninsured at diagnosis. The most common cancers were hematologic (17%), female genital (16%), thyroid (11%), and breast (11%). Metastasis was observed at diagnosis for 24% of AYAs. The expected 5-year survival for the safety-net population based on SEER data was 77%, whereas observed survival was 68%, which equated to an absolute survival difference of 9.5% (95% CL: 5.3%, 14%) after 5 years. Conclusions: Our results suggest that survival estimates of AYAs with cancer in SEER may not be generalizable to AYAs with cancer at an urban safety-net institution in Texas. Lack of generalizability may be relevant to other populations not represented in SEER. The reasons for poor survival of AYAs with cancer at safety-net institutions should be explored.

Trends in cause of death among patients with multiple myeloma in Puerto Rico and the United States SEER population, 1987-2013.

Multiple myeloma (MM) survival has improved due to recent developments in MM treatment. As a result, other co-morbid conditions may be of increasing importance to MM patients' long-term survival. This study examines trends in common causes of death among patients with MM in Puerto Rico, and in the US Surveillance, Epidemiology, and End Results (SEER) population. We analyzed the primary cause of death among incident MM cases recorded in the Puerto Rico Central Cancer Registry (n = 3,018) and the US SEER Program (n = 67,733) between 1987 and 2013. We calculated the cumulative incidence of death due to the eight most common causes and analyzed temporal trends in mortality rates using joinpoint regression. Analyses of SEER were also stratified by Hispanic ethnicity. MM accounted for approximately 72% of all reported deaths among persons diagnosed with MM in Puerto Rico and in SEER. In both populations, the proportion of patients who died from MM decreased with increasing time since diagnosis. Age-standardized temporal trends showed a decreased MM-specific mortality rate among US SEER (annual percent change [APC] = -5.0) and Puerto Rican (APC = -1.8) patients during the study period, and particularly after 2003 in non-Hispanic SEER patients. Temporal decline in non-MM causes of death was also observed among US SEER (APC = -2.1) and Puerto Rican (APC = -0.1) populations. MM-specific mortality decreased, yet remained the predominant cause of death for individuals diagnosed with MM over a 26-year period. The most pronounced decreases in MM-specific death occurred after 2003, which suggests a possible influence of more recently developed MM therapies.

Trends of incidence and survival of patients with chronic myelomonocytic leukemia between 1999 and 2014: A comparison between Swiss and American population-based cancer registries

BackgroundChronic myelomonocytic leukemia (CMML) is a rare hematopoietic malignancy. Treatment with hypomethylating agents (HMA) was introduced between 2004 and 2006 but its impact on population-based survival remains controversial. The aim of this study was to investigate epidemiological characteristics and survival before and after introduction of HMA treatment. MethodsWe performed a population-based analysis of CMML cases reported to the Cantonal Cancer Registries in Switzerland (SWISS) and the Surveillance, Epidemiology, and End Results (SEER) Program from the United States for 1999–2006 (before HMA) and 2007–2014 (after HMA). Time trends were compared for these two time periods. Results423 and 4144 new CMML cases were reported to the SWISS and SEER registries, respectively. We observed an increasing proportion of older patients ≥75 years in the SWISS (50.3%–62.3%) compared to a decreasing one in the SEER population (59.1%–55.1%). Age standardized incidence-rates were similar and remained stable in both countries (0.32–0.38 per 100’000 py). Relative survival (RS) improved significantly in the SEER (3 years 27%–37%; 5 years 19%–23%; p < 0.001 for both) but remained stable in the SWISS population (3 years 48% to 40%; 5 years 34% to 26%; n.s. for both). ConclusionsWith the exception of opposing age-trends, epidemiologic characteristics are similar in both countries and comparable to other population-based registries. RS remains poor and different time trends of population-based survival cannot be faithfully explained by HMA but most likely by changes in diagnostic accuracy within prognostically distinct age-groups.

Cancer risk in microscopic colitis: a retrospective cohort study

BackgroundThe long-term natural history of microscopic colitis (MC) (collagenous colitis (CC), lymphocytic colitis (LC)), traditionally considered relapsing but non-progressive diseases, is poorly defined. Whether persistent histologic inflammation in such diseases is associated with an increased risk of colorectal neoplasia (CRN) or extracolonic cancers has not been robustly established.MethodsThis retrospective cohort included diagnosed with MC at a referral center. Rates of CRN and extracolonic cancer were compared to patients undergoing screening colonoscopy (n = 306) and to the United States population using data from the Surveillance, Epidemiology, and End-Results (SEER) program. Standardized incidence ratios (SIR) and 95% confidence intervals were calculated and multivariable regression models used to identify the effect of MC diagnosis and severity on cancer risk.ResultsOur study included 221 patients with microscopic colitis (112 CC, 109 LC) among whom 77% were women. Compared to the colonoscopy control population, MC was associated with similar odds of tubular adenoma (Odds ratio (OR) 1.07, 95% CI 0.69–1.66) or villous adenoma (OR 1.26, 95% CI 0.17–9.42). Compared to patients with a single episode of MC, those with 2 or more episodes had similar risk of colon cancer (OR 0.83, 95% CI 0.20–3.39) or tubular adenoma (OR 1.49 95% CI 0.83–2.67). We also identified no statistical increase in the rates of cancer in the MC population compared to US-SEER data.ConclusionMicroscopic colitis was not associated with increased risk of CRN and extracolonic cancers when compared to controls undergoing colonoscopy or the US SEER population.