A Systematic Review of Estimating Breast Cancer Recurrence at the Population Level With Administrative Data

Prognostic model for predicting recurrence in breast cancer patients in Saudi Arabia

Breast cancer recurrence presents a significant global health challenge, and accurate prediction is crucial for effective patient management and improved outcomes. Reliable predictive tools can help tailor therapeutic approaches, provide personalized care, and enhance patient outcomes. In light of the current lack of such tools in clinical practice, our study aimed to develop predictive models for breast cancer recurrence within three years of treatment. We analyzed data from 408 breast cancer patients at the King Fahd Specialist Hospital in Dammam, Saudi Arabia and divided them into training (n = 285) and test (n = 123) cohorts. Using multivariable penalized logistic regression combined with a nested cross-validation framework and multivariate Cox regression analysis to determine time-dependent risk factors for breast cancer recurrence, we developed prognostic models that incorporated age, stage, tumor size, and treatment type. We evaluated the performance of the models using both the area under the receiver operating characteristic curve for multivariate logistic regression and C-index for multivariate Cox regression. The multivariate logistic regression model achieved an area under the curve (AUC) of 76% (95% confidence interval [CI]: 72-81%) for the training set and 76% (95% CI: 66-87%) for the test set. The Cox regression analysis yielded a C-index of 0.81 for the training set (95% CI: 0.73-0.84) and 0.84 for the test set (95% CI: 0.76-0.89). Chemotherapy was found to decrease recurrence odds by 86% (adjusted odds ratio [AOR]: 0.143, 95% CI: 0.089-0.218, p < 0.0001), and surgery resulted in a 99% reduction in recurrence probability (AOR: 0.009, 95% CI: 0.005-0.014, p < 0.0001). Increased tumor size improved the recurrence odds by 48.5% (AOR: 1.485, 95% CI: 1.128-1.918, p = 0.0043), while age did not significantly predict recurrence (AOR: 0.841, 95% CI: 0.657-1.061, p = 0.1398). The newly developed, routinely collected baseline clinical features to predict breast cancer recurrence may be a valuable tool for clinical decision-making and is freely available online. The tool can be accessed through the following link: https://iv3p9h-nurudeen-adegoke.shinyapps.io/breast_cancer .

A population-based estimation of breast cancer recurrence in northeast Italy with administrative healthcare databases

BACKGROUND/AIM

Information on the long-term frequency of recurrence is of paramount importance for the increasing number of women living several years after breast cancer (BC) diagnosis and for their caregivers. The study aims to estimate the cumulative incidence of recurrence until 10 years after diagnosis in Italian women diagnosed with BC using population-based cancer registries.

METHODS

Women diagnosed with stage I to III BC during 2004-2010 from Friuli Venezia Giulia and Veneto (Italy) cancer registries were included (n = 5825). Recurrence status after a disease-free period was ascertained through individual-level linked databases using treatment or procedure codes from claims. Cumulative incidence of recurrence was calculated in the presence of competing risks (second cancer or death).

RESULTS

During a median follow-up of 13.5 years, 1522 out of 5825 women experienced a recurrence with an estimated 10-years cumulative incidence of 20.8 % (95 %CI:19.7-21.8 %), decreasing from 23.7 % in 2004-2006 to 18.5 % in 2007-2010. Women younger than 40 years (40.5 %), with stage III (41.8 %) and triple-negative BC (32.5 %) showed a higher 10-year incidence of recurrence. At 10 years after a BC diagnosis, 83.9 % of women were alive: 67.5 % without any cancer-related events, 12.4 % after recurrence and 4.0 % after second primary cancer. 10-years survival was higher than 90 % for women with stage I BC and 58.1 % for those with stage III (3.2 % and 27.3 % deaths after recurrence, respectively).

DISCUSSION

This Italian study provide detailed population-based information on the incidence of recurrence and other outcomes after BC and may be replicated in other Italian and European areas.

Metastatic Breast Cancer Prevalence in New South Wales, Australia, in 2016: A Health Record Linkage Study

AIM

To estimate the number of females living with metastatic breast cancer (MBC) in New South Wales (NSW), Australia, in 2016 using linked health records.

METHODS

The primary study dataset (cohort 1) included females in the NSW Cancer Registry (NSWCR) with breast cancer diagnosed during 2001-2002 and 2006-2007 linked with administrative hospital records, medicine dispensing, radiation services, and death records. From this dataset we counted the number with a record of de novo MBC or recurrent MBC (following stage I-III cancer) alive at the end of each year (2001-2015). The second dataset (cohort 2) included females with breast cancer diagnosed 2003-2005 and 2008-2015 without linked records. We imputed MBC prevalence for cohort 2 by calculating MBC prevalence proportions at the end of each year in cohort 1 and applying these proportions to NSWCR incidence counts in cohort 2.

RESULTS

Cohort 1 comprised 16,521 females with breast cancer, of whom 4364 had MBC recorded (976 de novo; 3388 recurrent). A total of 1245 individuals with MBC recorded were alive on January 1, 2016 (270 de novo, 21.7%; 975 recurrent, 78.3%). When extrapolated to all females diagnosed with breast cancer in 2001-2015 in NSW, 5009 individuals were estimated to be living with MBC on January 1, 2016 (1609 de novo, 32.1%; 3400 recurrent, 67.9%).

CONCLUSION

This study estimates that a large number of individuals are living with MBC and demonstrates the importance of identifying individuals with recurrent MBC, in addition to de novo MBC, to inform funding and delivery of appropriate clinical and supportive care services.

Incidence and temporal patterns of true recurrences and second primaries in women with breast cancer: A 10-year competing risk-adjusted analysis

INTRODUCTION

We report a population-based, competing risk-adjusted analysis of the risk and timing of true recurrences and second primaries in women with breast cancer (BC), that are still ill-defined.

METHODS

We performed a manual review of medical charts of 1988 BC patients from a cancer registry in northern Italy (2000-2013). The occurrence and timing of true recurrences (TRs, including local, regional and distant recurrences) and second BCs (SBCs, including ipsilateral and contralateral SBC) during 10 years of follow-up were evaluated. The prognostic factors for TRs and SBCs were identified using the Fine and Gray model.

RESULTS

The cumulative incidence was 13.7 % (95 % confidence interval (CI), 12.2-15.3 %) for TRs and 4.6 % (95 % CI, 3.7-5.7 %) for SBCs. The median time to detection varied between 3.4 (TRs) and 5.1 (SBCs) years. The risk of TRs had two peaks, one between the 2nd and the 3rd year of follow-up and another between the 7th and the 8th year. The subhazard of SBCs fluctuated for five years, had a drop between the 6th and the 7th year and a marked peak between the 8th and the 9th year. Prognostic factors for TRs (tumour stage and grade, lymph node status and residual disease) and SBCs (patient age and -inverse association- hormone therapy) were different. In the 9th-10th year of follow-up, the excess incidence of total BC episodes as compared with the expected incidence of BC was no longer significant (standardised incidence ratio, 1.15; 95 % CI, 0.86-1.53).

CONCLUSIONS

The multifaceted results of this study warrant further research into the risk and timing of all types of BC recurrence.

Misclassification of second primary and recurrent breast cancer in the surveillance epidemiology and end results registry

The Surveillance Epidemiology and End Results (SEER) registry incorporates laterality, histology, latency, and topography to identify second primary breast cancers. Contralateral tumors are classified as second primaries, but ipsilaterals are subject to additional inclusion criteria that increase specificity but may induce biases. It is important to understand how classification methods affect accuracy of second tumor classification. We collected estrogen, progesterone, and human epidermal growth factor receptor 2 (ER, PR, Her2) status for 11,838 contralateral and 5,371 ipsilateral metachronous secondary tumors and estimated concordance odds ratios (cORs) to evaluate receptor dependence (the tendency for tumors to share receptor status) by laterality. If only second primaries are included, receptor dependence should be similar for contralateral and ipsilateral tumors. Thus, we compared ratios of cORs as a measure of inaccuracy. Cases who met ipsilateral second primary criteria were younger and had less aggressive primary tumor characteristics compared to contralateral tumors. Time to secondary tumors was (by definition) longer for ipsilaterals than contralaterals, especially among ER + primaries. Overall and in multiple strata, ipsilateral tumors showed higher receptor dependence than contralateral tumors (ratios of cORs > 1), suggesting some SEER-included ipsilaterals are recurrences. SEER multiple primary criteria increase specificity, but remain inaccurate and may lack sensitivity. The dearth of early occurring ipsilateral tumors (by definition), coupled with high observed receptor dependence among ipsilaterals, suggests important inaccuracies. Datasets that allow comparison of pathologist- and SEER-classification to true multi-marker genomic dependence are needed to understand inaccuracies induced by SEER definitions.

Retrospective Case-Cohort Study on Risk Factors for Developing Distant Metastases in Women With Breast Cancer

OBJECTIVE

This study aimed to identify risk factors associated with the development of metastases in breast cancer patients, to investigate survival rates, and the relationship between local recurrences and distant metastases.

METHODS

This retrospective case-cohort study included women with breast cancer who were treated at a certified Breast Unit between 2001 and 2015. Cases who developed distant metastases were compared to controls based on diagnosis year, stage, and age at diagnosis. Comprehensive information on patient characteristics, tumor biology, and treatment options was gathered.

RESULTS

The study included 412 patients who developed distant metastases and 433 controls who remained metastasis-free over a median follow-up of 150 months (interquartile range 87-202). The 20-year overall survival was 99.23% for the control group and 23.62% for those with metastasis (p < 0.01). Significant risk factors for metastasis included lobular invasive carcinoma (odds ratio (OR) 2.26, p < 0.001), triple-negative subtype (OR 4.06, p = 0.002), high tumor grade (OR 2.62, p = 0.004), larger tumor size (OR 1.02, p < 0.001), lymph node involvement (p < 0.001), and loco-regional recurrence (OR 4.32, p < 0.001). Progesterone receptor (PR) expression was protective (OR 0.52, 95% confidence interval 0.34-0.81, p = 0.003). Machine learning models supported these findings, though their clinical significance was limited.

CONCLUSIONS

Lobular invasive carcinoma, specific tumor subtypes, high grade, large tumor size, lymph node involvement, and loco-regional recurrence are all significant risk factors for distant metastasis, whereas PR expression is protective. The potential of machine learning in predicting metastasis was explored, showing promise for future personalized risk assessment.

Estimated Incidence and Prevalence of Metastatic Breast Cancer in Northern Ireland, 2009 to 2020

This cohort study examines the incidence and prevalence of metastatic breast cancer in Northern Ireland using population-based cancer registry data and health records.

Recurrence and associated factors of breast cancer among women in sub saharan africa, systematic review and meta-analysis

BACKGROUND

Recurrence of breast cancer is a critical indicator of disease progression and survival rates. Therefore, this systematic review and meta-analysis aimed to assess the prevalence of recurrence and associated factors of breast cancer in Sub Saharan Africa.

METHODS

We conducted a thorough search of the following international databases between January 1 and February 7, 2024: PubMed, EMBASE, CINAHL, Google Scholar, Science Direct, and the Cochrane Library. A data extraction format was used by two authors to independently extract all required data. STATA Version 14 was used to evaluate the quantitative data and the Cochrane Q test statistics and I2 test were used to evaluate the heterogeneity among the included studies using a random effects meta-analysis model.

RESULTS

The overall prevalence of breast cancer recurrence is 27.44 % (95%CI: 26.41, 28.46). The highest prevalence was found in Uganda (89.927 % (87.0, 92.851)), followed by Tanzania (82.174 % (77.228, 87.120)). Involved deep surgical margin (OR: 3.62, 95 % CI: 2.11, 5.12), positive lymph node status (OR: 6.85, 95 % CI: 1.42, 12.3)), and histological grade III (OR: 7.43, 95 % CI: 3.56, 11.3)) were all significantly associated factors.

CONCLUSION

The pooled prevalence of breast cancer recurrence in this review was significantly high. Histological grade III, positive lymph node statuses, clinical staging III, and involved deep surgical margin were associated factors. Therefore, frequent monitoring, regular screenings, imaging tests, and consultations with oncologists, take extra care to ensure clear and deep surgical margins using advanced imaging techniques are highly recommended.

Racial and Ethnic Differences in Psychosocial Care Use Among Adults With Metastatic Breast Cancer: A Retrospective Analysis Across Six New York City Health Systems

PURPOSE

A metastatic breast cancer (mBC) diagnosis can affect physical and emotional well-being. However, racial and ethnic differences in receipt of outpatient psychosocial care and supportive care medications in adults with mBC are not well described.

METHODS

Adults with mBC were identified in the INSIGHT-Clinical Research Network, a database inclusive of >12 million patients receiving care across six New York City health systems. Outpatient psychosocial care was operationalized using Common Procedure Terminology codes for outpatient psychotherapy or counseling. Psychosocial/supportive care medications were defined using Rx Concept Unique Identifier codes. Associations between race/ethnicity and outpatient care and medication use were evaluated using logistic regression.

RESULTS

Among 5,429 adults in the analytic cohort, mean age was 61 years and <1% were male; 53.6% were non-Hispanic White (NHW), 21.4% non-Hispanic Black (NHB), 15.9% Hispanic, 6.1% Asian/Native Hawaiian/Pacific Islander (A/NH/PI), and 3% other or unknown. Overall, 4.1% had ≥one outpatient psychosocial care visit and 63.4% were prescribed ≥one medication. Adjusted for age, compared with NHW, Hispanic patients were more likely (odds ratio [OR], 2.14 [95% CI, 1.55 to 2.92]) and A/NH/PI patients less likely (OR, 0.35 [95% CI, 0.12 to 0.78]) to have an outpatient visit. NHB (OR, 0.59 [95% CI, 0.51 to 0.68]) and Asian (OR, 0.36 [95% CI, 0.29 to 0.46]) patients were less likely to be prescribed medications.

CONCLUSION

Despite the prevalence of depression, anxiety, and distress among patients with mBC, we observed low utilization of psychosocial outpatient care. Supportive medication use was more prevalent, although differences observed by race/ethnicity suggest that unmet needs exist.

Evaluation of Algorithms Using Automated Health Plan Data to Identify Breast Cancer Recurrences

BACKGROUND

We updated algorithms to identify breast cancer recurrences from administrative data, extending previously developed methods.

METHODS

In this validation study, we evaluated pairs of breast cancer recurrence algorithms (vs. individual algorithms) to identify recurrences. We generated algorithm combinations that categorized discordant algorithm results as no recurrence [High Specificity and PPV (positive predictive value) Combination] or recurrence (High Sensitivity Combination). We compared individual and combined algorithm results to manually abstracted recurrence outcomes from a sample of 600 people with incident stage I-IIIA breast cancer diagnosed between 2004 and 2015. We used Cox regression to evaluate risk factors associated with age- and stage-adjusted recurrence rates using different recurrence definitions, weighted by inverse sampling probabilities.

RESULTS

Among 600 people, we identified 117 recurrences using the High Specificity and PPV Combination, 505 using the High Sensitivity Combination, and 118 using manual abstraction. The High Specificity and PPV Combination had good specificity [98%, 95% confidence interval (CI): 97-99] and PPV (72%, 95% CI: 63-80) but modest sensitivity (64%, 95% CI: 44-80). The High Sensitivity Combination had good sensitivity (80%, 95% CI: 49-94) and specificity (83%, 95% CI: 80-86) but low PPV (29%, 95% CI: 25-34). Recurrence rates using combined algorithms were similar in magnitude for most risk factors.

CONCLUSIONS

By combining algorithms, we identified breast cancer recurrences with greater PPV than individual algorithms, without additional review of discordant records.

IMPACT

Researchers should consider tradeoffs between accuracy and manual chart abstraction resources when using previously developed algorithms. We provided guidance for future studies that use breast cancer recurrence algorithms with or without supplemental manual chart abstraction.

Decline in the Incidence of Distant Recurrence of Breast Cancer: A Population-Based Health Record Linkage Study, Australia 2001-2016

BACKGROUND

We investigated differences in cumulative incidence of first distant recurrence (DR) following non-metastatic breast cancer over a time period when new adjuvant therapies became available in Australia.

METHODS

We conducted a health record linkage study of females with localized (T1-3N0) or regional (T4 or N+) breast cancer in the New South Wales Cancer Registry in 2001 to 2002 and 2006 to 2007. We linked cancer registry records with administrative records from hospitals, dispensed medicines, radiotherapy services, and death registrations to estimate the 9-year cumulative incidence of DR and describe use of adjuvant treatment.

RESULTS

The study included 13,170 women (2001-2002 n = 6,338, 2006-2007 n = 6,832). The 9-year cumulative incidence of DR was 3.6% [95% confidence interval (CI), 2.3%-4.9%] lower for 2006-2007 diagnoses (15.0%) than 2001-2002 (18.6%). Differences in the annual hazard of DR between cohorts were largest in year two. DR incidence declined for localized and regional disease. Decline was largest for ages <40 years (absolute difference, 14.4%; 95% CI, 8.3%-20.6%), whereas their use of adjuvant chemotherapy (2001-2002 49%, 2006-2007 75%) and HER2-targeted therapy (2001-2002 0%, 2006-2007 16%) increased. DR did not decline for ages ≥70 years (absolute difference, 0.9%; 95% CI, -3.6%-1.8%) who had low use of adjuvant chemotherapy and HER2-targeted therapy.

CONCLUSIONS

This whole-of-population study suggests that DR incidence declined over time. Decline was largest for younger ages, coinciding with changes to adjuvant breast cancer therapy.

IMPACT

Study findings support the need for trials addressing questions relevant to older people and cancer registry surveillance of DR to inform cancer control programs.

Extracting cancer concepts from clinical notes using natural language processing: a systematic review

BACKGROUND

Extracting information from free texts using natural language processing (NLP) can save time and reduce the hassle of manually extracting large quantities of data from incredibly complex clinical notes of cancer patients. This study aimed to systematically review studies that used NLP methods to identify cancer concepts from clinical notes automatically.

METHODS

PubMed, Scopus, Web of Science, and Embase were searched for English language papers using a combination of the terms concerning "Cancer", "NLP", "Coding", and "Registries" until June 29, 2021. Two reviewers independently assessed the eligibility of papers for inclusion in the review.

RESULTS

Most of the software programs used for concept extraction reported were developed by the researchers (n = 7). Rule-based algorithms were the most frequently used algorithms for developing these programs. In most articles, the criteria of accuracy (n = 14) and sensitivity (n = 12) were used to evaluate the algorithms. In addition, Systematized Nomenclature of Medicine-Clinical Terms (SNOMED-CT) and Unified Medical Language System (UMLS) were the most commonly used terminologies to identify concepts. Most studies focused on breast cancer (n = 4, 19%) and lung cancer (n = 4, 19%).

CONCLUSION

The use of NLP for extracting the concepts and symptoms of cancer has increased in recent years. The rule-based algorithms are well-liked algorithms by developers. Due to these algorithms' high accuracy and sensitivity in identifying and extracting cancer concepts, we suggested that future studies use these algorithms to extract the concepts of other diseases as well.

Machine Learning Algorithm to Estimate Distant Breast Cancer Recurrence at the Population Level with Administrative Data

Purpose

High-quality population-based cancer recurrence data are scarcely available, mainly due to complexity and cost of registration. For the first time in Belgium, we developed a tool to estimate distant recurrence after a breast cancer diagnosis at the population level, based on real-world cancer registration and administrative data.

Methods

Data on distant cancer recurrence (including progression) from patients diagnosed with breast cancer between 2009-2014 were collected from medical files at 9 Belgian centers to train, test and externally validate an algorithm (i.e., gold standard). Distant recurrence was defined as the occurrence of distant metastases between 120 days and within 10 years after the primary diagnosis, with follow-up until December 31, 2018. Data from the gold standard were linked to population-based data from the Belgian Cancer Registry (BCR) and administrative data sources. Potential features to detect recurrences in administrative data were defined based on expert opinion from breast oncologists, and subsequently selected using bootstrap aggregation. Based on the selected features, classification and regression tree (CART) analysis was performed to construct an algorithm for classifying patients as having a distant recurrence or not.

Results

A total of 2507 patients were included of whom 216 had a distant recurrence in the clinical data set. The performance of the algorithm showed sensitivity of 79.5% (95% CI 68.8-87.8%), positive predictive value (PPV) of 79.5% (95% CI 68.8-87.8%), and accuracy of 96.7% (95% CI 95.4-97.7%). The external validation resulted in a sensitivity of 84.1% (95% CI 74.4-91.3%), PPV of 84.1% (95% CI 74.4-91.3%), and an accuracy of 96.8% (95% CI 95.4-97.9%).

Conclusion

Our algorithm detected distant breast cancer recurrences with an overall good accuracy of 96.8% for patients with breast cancer, as observed in the first multi-centric external validation exercise.

A Validated Register-Based Algorithm to Identify Patients Diagnosed with Recurrence of Surgically Treated Stage I Lung Cancer in Denmark

Introduction

Recurrence of cancer is not routinely registered in Danish national health registers. This study aimed to develop and validate a register-based algorithm to identify patients diagnosed with recurrent lung cancer and to estimate the accuracy of the identified diagnosis date.

Material and Methods

Patients with early-stage lung cancer treated with surgery were included in the study. Recurrence indicators were diagnosis and procedure codes recorded in the Danish National Patient Register and pathology results recorded in the Danish National Pathology Register. Information from CT scans and medical records served as the gold standard to assess the accuracy of the algorithm.

Results

The final population consisted of 217 patients; 72 (33%) had recurrence according to the gold standard. The median follow-up time since primary lung cancer diagnosis was 29 months (interquartile interval: 18-46). The algorithm for identifying a recurrence reached a sensitivity of 83.3% (95% CI: 72.7-91.1), a specificity of 93.8% (95% CI: 88.5-97.1), and a positive predictive value of 87.0% (95% CI: 76.7-93.9). The algorithm identified 70% of the recurrences within 60 days of the recurrence date registered by the gold standard method. The positive predictive value of the algorithm decreased to 70% when the algorithm was simulated in a population with a recurrence rate of 15%.

Conclusion

The proposed algorithm demonstrated good performance in a population with 33% recurrences over a median of 29 months. It can be used to identify patients diagnosed with recurrent lung cancer, and it may be a valuable tool for future research in this field. However, a lower positive predictive value is seen when applying the algorithm in populations with low recurrence rates.

Validated algorithms for identifying timing of second event of oropharyngeal squamous cell carcinoma using real-world data

BACKGROUND

Understanding occurrence and timing of second events (recurrence and second primary cancer) is essential for cancer specific survival analysis. However, this information is not readily available in administrative data.

METHODS

Alberta Cancer Registry, physician claims, and other administrative data were used. Timing of second event was estimated based on our developed algorithm. For validation, the difference, in days between the algorithm estimated and the chart-reviewed timing of second event. Further, the result of Cox-regression modeling cancer-free survival was compared to chart review data.

RESULTS

Majority (74.3%) of the patients had a difference between the chart-reviewed and algorithm-estimated timing of second event falling within the 0-60 days window. Kaplan-Meier curves generated from the estimated data and chart review data were comparable with a 5-year second-event-free survival rate of 75.4% versus 72.5%.

CONCLUSION

The algorithm provided an estimated timing of second event similar to that of the chart review.

Analysis Predictors of the Outcome of Adjuvant of Hormone Therapy on Estrogen Receptor-positive Breast Cancer in Indonesia

BACKGROUND: An existing study reported variation of the outcome of adjuvant hormone therapy on breast cancer. AIM: This study aimed to examine predictors of the hormone therapy to the outcome of recurrence-free survival (RFS) of estrogen receptor-positive (ER+) breast cancer. METHODS: In this cohort study, we followed up 219 eligible breast cancer patients with ER+ who had hormone therapy in 2017–2018. Age of patients, cancer stage, and various histopathology parameters were collected from the medical records, then we followed up with the patients within 2 years (2019–2020) to assess the RFS outcome. Bivariate analysis was conducted to assess the association between the clinicopathology parameters with RFS outcome. Multivariate analysis with logistic regression was also performed to see the dominant predictor. Mediation path analysis was also performed to determine the estimated effect of a predictor on the level of RFS and to see the visualization of the association of predictors with RFS. RESULTS: Breast cancer RFS was 91.3% within 2 years of hormone therapy. The recurrent rate was only 8.7%, which most of them (68.4%) were local. There was no association of age, lymphovascular invasion (LVI), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status with RFS. Based on the molecular subtype, the RFS was better in luminal A (p = 0.045), and also better gradually in the lower stage (p = 0.001). Multivariate analysis shows that the cancer stage was the dominant predictor of the RFS outcome (p = 0.001) with OR = 4.271 (Exp[B] = 1.937–9.417). Mediation analysis also found that there was a positively associated molecular subtype with RFS through cancer stage mediation (r = 16.7%, p = 0.006) but no statistically significant association of age, LVI, PR, and HER2 status (p > 0.005). CONCLUSION: Cancer stage is the main predictor of RFS of hormone therapy outcome. Luminal A is most also likely to have a better outcome of RFS, especially mediated by the lower stage.

BRENDA-Score, a Highly Significant, Internally and Externally Validated Prognostic Marker for Metastatic Recurrence: Analysis of 10,449 Primary Breast Cancer Patients

Simple Summary

The BRENDA-Score provides an easy to use tool for clinicians to estimate the risk of recurrence in primary breast cancer. The algorithm has been validated via a second independent database and provides five recurrence risk groups. This grouping helps clinicians to encourage high risk patients to undergo the recommended treatment.

Abstract

Background Current research in breast cancer focuses on individualization of local and systemic therapies with adequate escalation or de-escalation strategies. As a result, about two-thirds of breast cancer patients can be cured, but up to one-third eventually develop metastatic disease, which is considered incurable with currently available treatment options. This underscores the importance to develop a metastatic recurrence score to escalate or de-escalate treatment strategies. Patients and methods Data from 10,499 patients were available from 17 clinical cancer registries (BRENDA-project. In total, 8566 were used to develop the BRENDA-Index. This index was calculated from the regression coefficients of a Cox regression model for metastasis-free survival (MFS). Based on this index, patients were categorized into very high, high, intermediate, low, and very low risk groups forming the BRENDA-Score. Bootstrapping was used for internal validation and an independent dataset of 1883 patients for external validation. The predictive accuracy was checked by Harrell’s c-index. In addition, the BRENDA-Score was analyzed as a marker for overall survival (OS) and compared to the Nottingham prognostic score (NPS). Results: Intrinsic subtypes, tumour size, grading, and nodal status were identified as statistically significant prognostic factors in the multivariate analysis. The five prognostic groups of the BRENDA-Score showed highly significant (p < 0.001) differences regarding MFS:low risk: hazard ratio (HR) = 2.4, 95%CI (1.7–3.3); intermediate risk: HR = 5.0, 95%CI.(3.6–6.9); high risk: HR = 10.3, 95%CI (7.4–14.3) and very high risk: HR = 18.1, 95%CI (13.2–24.9). The external validation showed congruent results. A multivariate Cox regression model for OS with BRENDA-Score and NPS as covariates showed that of these two scores only the BRENDA-Score is significant (BRENDA-Score p < 0.001; NPS p = 0.447). Therefore, the BRENDA-Score is also a good prognostic marker for OS. Conclusion: The BRENDA-Score is an internally and externally validated robust predictive tool for metastatic recurrence in breast cancer patients. It is based on routine parameters easily accessible in daily clinical care. In addition, the BRENDA-Score is a good prognostic marker for overall survival. Highlights: The BRENDA-Score is a highly significant predictive tool for metastatic recurrence of breast cancer patients. The BRENDA-Score is stable for at least the first five years after primary diagnosis, i.e., the sensitivities and specificities of this predicting system is rather similar to the NPI with AUCs between 0.76 and 0.81 the BRENDA-Score is a good prognostic marker for overall survival.

Weakly supervised temporal model for prediction of breast cancer distant recurrence

Efficient prediction of cancer recurrence in advance may help to recruit high risk breast cancer patients for clinical trial on-time and can guide a proper treatment plan. Several machine learning approaches have been developed for recurrence prediction in previous studies, but most of them use only structured electronic health records and only a small training dataset, with limited success in clinical application. While free-text clinic notes may offer the greatest nuance and detail about a patient’s clinical status, they are largely excluded in previous predictive models due to the increase in processing complexity and need for a complex modeling framework. In this study, we developed a weak-supervision framework for breast cancer recurrence prediction in which we trained a deep learning model on a large sample of free-text clinic notes by utilizing a combination of manually curated labels and NLP-generated non-perfect recurrence labels. The model was trained jointly on manually curated data from 670 patients and NLP-curated data of 8062 patients. It was validated on manually annotated data from 224 patients with recurrence and achieved 0.94 AUROC. This weak supervision approach allowed us to learn from a larger dataset using imperfect labels and ultimately provided greater accuracy compared to a smaller hand-curated dataset, with less manual effort invested in curation.

Predicting breast cancer 5-year survival using machine learning: A systematic review

BACKGROUND

Accurately predicting the survival rate of breast cancer patients is a major issue for cancer researchers. Machine learning (ML) has attracted much attention with the hope that it could provide accurate results, but its modeling methods and prediction performance remain controversial. The aim of this systematic review is to identify and critically appraise current studies regarding the application of ML in predicting the 5-year survival rate of breast cancer.

METHODS

In accordance with the PRISMA guidelines, two researchers independently searched the PubMed (including MEDLINE), Embase, and Web of Science Core databases from inception to November 30, 2020. The search terms included breast neoplasms, survival, machine learning, and specific algorithm names. The included studies related to the use of ML to build a breast cancer survival prediction model and model performance that can be measured with the value of said verification results. The excluded studies in which the modeling process were not explained clearly and had incomplete information. The extracted information included literature information, database information, data preparation and modeling process information, model construction and performance evaluation information, and candidate predictor information.

RESULTS

Thirty-one studies that met the inclusion criteria were included, most of which were published after 2013. The most frequently used ML methods were decision trees (19 studies, 61.3%), artificial neural networks (18 studies, 58.1%), support vector machines (16 studies, 51.6%), and ensemble learning (10 studies, 32.3%). The median sample size was 37256 (range 200 to 659820) patients, and the median predictor was 16 (range 3 to 625). The accuracy of 29 studies ranged from 0.510 to 0.971. The sensitivity of 25 studies ranged from 0.037 to 1. The specificity of 24 studies ranged from 0.008 to 0.993. The AUC of 20 studies ranged from 0.500 to 0.972. The precision of 6 studies ranged from 0.549 to 1. All of the models were internally validated, and only one was externally validated.

CONCLUSIONS

Overall, compared with traditional statistical methods, the performance of ML models does not necessarily show any improvement, and this area of research still faces limitations related to a lack of data preprocessing steps, the excessive differences of sample feature selection, and issues related to validation. Further optimization of the performance of the proposed model is also needed in the future, which requires more standardization and subsequent validation.

Identification of endometrial cancer recurrence - a validated algorithm based on nationwide Danish registries

INTRODUCTION

Recurrence of endometrial cancer is not routinely registered in the Danish national health registers. The aim of this study was to develop and validate a register-based algorithm to identify women diagnosed with endometrial cancer recurrence in Denmark to facilitate register-based research in this field.

MATERIAL AND METHODS

We conducted a cohort study based on data from Danish health registers. The algorithm was designed to identify women with recurrence and estimate the accompanying diagnosis date, which was based on information from the Danish National Patient Registry and the Danish National Pathology Registry. Indicators of recurrence were pathology registrations and procedure or diagnosis codes suggesting recurrence and related treatment. The gold standard for endometrial cancer recurrence originated from a Danish nationwide study of 2612 women diagnosed with endometrial cancer, FIGO stage I-II during 2005-2009. Recurrence was suspected in 308 women based on pathology reports, and recurrence suspicion was confirmed or rejected in the 308 women based on reviews of the medical records. The algorithm was validated by comparing the recurrence status identified by the algorithm and the recurrence status in the gold standard.

RESULTS

After relevant exclusions, the final study population consisted of 268 women, hereof 160 (60%) with recurrence according to the gold standard. The algorithm displayed a sensitivity of 91.3% (95% confidence interval (CI): 85.8-95.1), a specificity of 91.7% (95% CI: 84.8-96.1) and a positive predictive value of 94.2% (95% CI: 89.3-97.3). The algorithm estimated the recurrence date within 30 days of the gold standard in 86% and within 60 days of the gold standard in 94% of the identified patients.

DISCUSSION

The algorithm demonstrated good performance; it could be a valuable tool for future research in endometrial cancer recurrence and may facilitate studies with potential impact on clinical practice.

A Validated Register-Based Algorithm to Identify Patients Diagnosed with Recurrence of Malignant Melanoma in Denmark

Purpose

Information on cancer recurrence is rarely available outside clinical trials. Wide exclusion criteria used in clinical trials tend to limit the generalizability of findings to the entire population of people living beyond a cancer disease. Therefore, population-level evidence is needed. The aim of this study was to develop and validate a register-based algorithm to identify patients diagnosed with recurrence after curative treatment of malignant melanoma.

Patients and Methods

Indicators of recurrence were diagnosis and procedure codes recorded in the Danish National Patient Register and pathology results recorded in the Danish National Pathology Register. Medical records on recurrence status and recurrence date in the Danish Melanoma Database served as the gold standard to assess the accuracy of the algorithm.

Results

The study included 1747 patients diagnosed with malignant melanoma; 95 (5.4%) were diagnosed with recurrence of malignant melanoma according to the gold standard. The algorithm reached a sensitivity of 93.7% (95% confidence interval (CI) 86.8–97.6), a specificity of 99.2% (95% CI: 98.6–99.5), a positive predictive value of 86.4% (95% CI: 78.2–92.4), and negative predictive value of 99.6% (95% CI: 99.2–99.9). Lin’s concordance correlation coefficient was 0.992 (95% CI: 0.989–0.996) for the agreement between the recurrence dates generated by the algorithm and by the gold standard.

Conclusion

The algorithm can be used to identify patients diagnosed with recurrence of malignant melanoma and to establish the timing of recurrence. This can generate population-level evidence on disease-free survival and diagnostic pathways for recurrence of malignant melanoma.