Supervised Machine Learning to Predict Follow-Up Among Adjuvant Endocrine Therapy Patients

2017 ◽  
Author(s):  
Morgan Harrell ◽  
Mia Levy ◽  
Daniel Fabbri
Keyword(s):  
Machine Learning ◽  
Endocrine Therapy ◽  
Adjuvant Endocrine Therapy ◽  
Supervised Machine Learning

scholarly journals FRI0046 PHARMACOGENOMICS-DRIVEN INDIVIDUALIZED PREDICTION OF TREATMENT RESPONSE TO METHOTREXATE IN PATIENTS WITH RHEUMATOID ARTHRITIS: A MACHINE LEARNING APPROACH

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 598.2-598
Author(s):  
E. Myasoedova ◽  
A. Athreya ◽  
C. S. Crowson ◽  
R. Weinshilboum ◽  
L. Wang ◽  
...  
Keyword(s):  
Rheumatoid Arthritis ◽  
Machine Learning ◽  
Supervised Machine Learning ◽  
Research Support ◽  
Eular Response ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Early Ra ◽  
Genome Wide

Background:Methotrexate (MTX) is the most common anchor drug for rheumatoid arthritis (RA), but the risk of missing the opportunity for early effective treatment with alternative medications is substantial given the delayed onset of MTX action and 30-40% inadequate response rate. There is a compelling need to accurately predicting MTX response prior to treatment initiation, which allows for effectively identifying patients at RA onset who are likely to respond to MTX.Objectives:To test the ability of machine learning approaches with clinical and genomic biomarkers to predict MTX response with replications in independent samples.Methods:Age, sex, clinical, serological and genome-wide association study (GWAS) data on patients with early RA of European ancestry from 647 patients (336 recruited in United Kingdom [UK]; 307 recruited across Europe; 70% female; 72% rheumatoid factor [RF] positive; mean age 54 years; mean baseline Disease Activity Score with 28-joint count [DAS28] 5.65) of the PhArmacogenetics of Methotrexate in RA (PAMERA) consortium was used in this study. The genomics data comprised 160 genome-wide significant single nucleotide polymorphisms (SNPs) with p<1×10-5 associated with risk of RA and MTX metabolism. DAS28 score was available at baseline and 3-month follow-up visit. Response to MTX monotherapy at the dose of ≥15 mg/week was defined as good or moderate by the EULAR response criteria at 3 months’ follow up visit. Supervised machine-learning methods were trained with 5-repeats and 10-fold cross-validation using data from PAMERA’s 336 UK patients. Class imbalance (higher % of MTX responders) in training was accounted by using simulated minority oversampling technique. Prediction performance was validated in PAMERA’s 307 European patients (not used in training).Results:Age, sex, RF positivity and baseline DAS28 data predicted MTX response with 58% accuracy of UK and European patients (p = 0.7). However, supervised machine-learning methods that combined demographics, RF positivity, baseline DAS28 and genomic SNPs predicted EULAR response at 3 months with area under the receiver operating curve (AUC) of 0.83 (p = 0.051) in UK patients, and achieved prediction accuracies (fraction of correctly predicted outcomes) of 76.2% (p = 0.054) in the European patients, with sensitivity of 72% and specificity of 77%. The addition of genomic data improved the predictive accuracies of MTX response by 19% and achieved cross-site replication. Baseline DAS28 scores and following SNPs rs12446816, rs13385025, rs113798271, and rs2372536 were among the top predictors of MTX response.Conclusion:Pharmacogenomic biomarkers combined with DAS28 scores predicted MTX response in patients with early RA more reliably than using demographics and DAS28 scores alone. Using pharmacogenomics biomarkers for identification of MTX responders at early stages of RA may help to guide effective RA treatment choices, including timely escalation of RA therapies. Further studies on personalized prediction of response to MTX and other anti-rheumatic treatments are warranted to optimize control of RA disease and improve outcomes in patients with RA.Disclosure of Interests:Elena Myasoedova: None declared, Arjun Athreya: None declared, Cynthia S. Crowson Grant/research support from: Pfizer research grant, Richard Weinshilboum Shareholder of: co-founder and stockholder in OneOme, Liewei Wang: None declared, Eric Matteson Grant/research support from: Pfizer, Consultant of: Boehringer Ingelheim, Gilead, TympoBio, Arena Pharmaceuticals, Speakers bureau: Simply Speaking

Impact of delays in initiation of adjuvant endocrine therapy and survival among patients with breast cancer.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. 537-537
Author(s):  
Kimberley Lee ◽  
Lisa K. Jacobs ◽  
Jodi Segal
Keyword(s):  
Breast Cancer ◽  
Overall Survival ◽  
Endocrine Therapy ◽  
Invasive Breast Cancer ◽  
Population Level ◽  
Initial Diagnosis ◽  
Adjuvant Endocrine Therapy ◽  
Delayed Surgery ◽  
Delayed Initiation

537 Background: Time to adjuvant endocrine therapy concerns patients and clinicians, but its impact on overall survival is not clear. There are no population level studies that address this question. Our primary objective is to describe the relationship between time from diagnosis of breast cancer to start of adjuvant endocrine therapy and overall survival. Methods: This is a population-based cohort study using prospectively collected population level data from the National Cancer Database (NCDB). The NCDB prospectively collects data on incident cancer cases from over 1500 Commission on Cancer-accredited facilities nationally. NCDB captures approximately 70% of incident cases of cancer in the United States. The participants are women with Stage II and III estrogen or progesterone receptor positive, human epidermal receptor 2 negative, invasive breast cancer who underwent definitive surgical treatment. Results: Of the 391,594 women in this study, 12,162 (3.1%) began treatment with adjuvant endocrine therapy more than 12 months after initial diagnosis of hormone receptor positive, invasive breast cancer. Mean age at diagnosis was 59.7 years (SD 13.4). Predictors of delayed initiation of adjuvant endocrine therapy include Black race or Hispanic ethnicity (adjusted odds ratio [aOR] of Black vs White, 1.57; 95% CI, 1.48-1.66; P < .001, Hispanic vs White, aOR 1.22, 95% CI 1.13-1.32; P < .001), Insurance other than private insurance (Medicare vs Private, aOR 1.09, 95% CI 1.01-1.17; P = .007, Medicaid vs Private, aOR 1.36, 95% CI 1.28-1.45; P < .001), higher stage of disease at diagnosis (Stage III vs II, aOR 1.24, 95% CI 1.19-1.30; P < .001), and delayed surgery or chemotherapy (Delayed surgery vs On-time lumpectomy, aOR 2.76, 95% CI 2.60-2.93; P < .001 and Delayed chemotherapy vs no chemotherapy, aOR 11.5, 95%CI 10.6-12.5). With median follow-up of 63.2 months, 67,335 (17.2%) patients died by the end of follow-up. Delayed initiation of AET resulted in no change in the hazard of death (HR, 1.00; 95% CI, 0.95-1.05; P = .97) compared to initiation within 12 months of diagnosis after adjusting for age, race and ethnicity, insurance type, urban vs rural residence, neighborhood income and education, comorbidity, cancer grade, stage, and receipt of timely or delayed surgery, chemotherapy, and/or radiation therapy. Conclusions: These results suggest that there may be no detriment to survival if initiation of adjuvant endocrine therapy occurs 12 to 24 months after initial diagnosis compared to within 12 months of diagnosis, as currently recommended.

Primary non-adherence to adjuvant endocrine therapy in older women with breast cancer.

2019 ◽  
Vol 37 (27_suppl) ◽  
pp. 285-285
Author(s):  
Vanina Tchuente ◽  
Donna Stern ◽  
Jaroslav Prchal ◽  
Judy Martin ◽  
Robyn Tamblyn ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Drug Use ◽  
Real Time ◽  
Endocrine Therapy ◽  
Surgical Care ◽  
Clinical Information ◽  
Adjuvant Endocrine Therapy ◽  
Administrative Claims ◽  
Patient Level

285 Background: Adjuvant endocrine therapy (AET) improves survival in hormone receptor positive breast cancer (HR+BC). Challenges with adherence to AET in seniors are well documented; however, there is limited knowledge on primary non-adherence (PNAD). PNAD is defined as non-initiation of a prescribed medication. Our aim is to characterize PNAD rates in women aged ≥ 65 with HR+BC and identify potential predictors, using real-time treatment information. Methods: Optimum is an e-health platform integrating real-time analysis of administrative claims data combined to patient-level clinical information on breast cancer. Optimum tracks care trajectories to identify deviations from best practice, using data from Quebec’s universal health insurance plan that covers all medical and pharmaceutical care. In this single-center feasibility study, we characterized PNAD as a non-initiation of AET within 10 days from the first prescription. Descriptive analyses were used to assess potential predictors. Results: Of the 57 patients enrolled, 9 were excluded due to lack of > 30 day follow up. In the remaining 48 patients, PNAD was 21 %. Baseline Charlson comorbidity index (0 vs 13 %), psychotropic drug use (20 % vs 26 %) and polypharmacy rate (10 % vs 11 %) were lower in PNAD patients, compared to primary-adherent patients. PNAD patients had larger average tumor size (1.8 cm vs 1.6 cm), more often overexpressing HER2NEU (10 % vs 3 %), more negative progesterone receptor (10 % vs 5 %). They also more often had lumpectomy (70 % vs 65 %), SLNB (70 % vs 58 %) and more frequent margin revisions (30 % vs 16 %). They more often received chemotherapy (30 % vs 0 %). At 30-day follow-up, 40 % of PNAD patients had not yet initiated AET. Conclusions: This study confirms the feasibility of combining real-time administrative data and patient-level clinical information to assess breast cancer quality care. PNAD in women with HR+BC was higher than expected. PNAD patients had less comorbidities and drug use, but more aggressive cancers and more often also had quality challenges with surgical care (margin revision). PNAD predictors can potentially be used to identify patients that may require additional support to optimize disease management.

scholarly journals Cardiac Diagnostic Feature and Demographic Identification Models: A Futuristic Approach for Smart Healthcare Using Machine Learning

2021 ◽  
Author(s):  
Deepak Kumar ◽  
Chaman Verma ◽  
Sanjay Dahiya ◽  
Pradeep Kumar Singh ◽  
Maria Simona Raboaca
Keyword(s):  
Machine Learning ◽  
Medical Records ◽  
Supervised Machine Learning ◽  
Age Group ◽  
Heart Patients ◽  
Kaplan Meier ◽  
And Gender ◽  
Clinical Records ◽  
The Impact

Around the world, every year, about 17 million people death cause happen due to CardioVascular Diseases (CVD). As per clinical records, primarily sufferers exhibit myocardial infarctions and Heart Failures (HF). Creatinine is a Musculo - skeletal waste product. The kidneys filter creatinine from the blood and excrete it through the urine in a healthy body. High creatinine levels can suggest renal problems. Elevated Serum Creatinine (SC) has been well established in the HF. Patients&rsquo; electronic medical records can be used to quantify symptoms and other related clinical laboratory test values, which would then be utilized to direct biostatistics exploration to uncover patterns and associations that doctors would otherwise miss. The latest American Heart Association guidelines for 1500 mg/d sodium tend to be sufficiently relevant for patients with stage A and B with HF. In this article, we used a dataset of the year 2015 of heart patients records of 299 patients. The present paper used the data analytic and statistical tools to verify the significant differences between alive and dead patients&rsquo; SC and Serum Sodium (SS). It also demonstrates the impact of significant features on abnormal SC and SS on the Survival-Status levels. The Age-Group feature, which is derived from age attribute and, Ejection Fraction (EF), anemia, platelets, Creatinine Phosphokinase (CPK), Blood-Pressure (BP), gender, diabetes, and smoking-status were utilized to determine the potential contributing features to mortality with Cox regression model. The Kaplan Meier plot was used to investigate the overall pattern of survival concerning age-group. During pre-processing of the dataset, Age and SS were removed due to multicollinear features during performing machine learning algorithms experiments. This paper also predicted patients&rsquo; survival, age group, and gender using supervised machine learning classifiers. Detection of significant features would help in making informed decisions to balance the lifestyle of heart patients. The author revealed that the patient&rsquo;s follow-up months, as well as SC, EF, CPK, and platelets, are sufficient key features to predict heart patient survival using Random Forest (RF) stratified 10-fold CV method with accuracy (96%) with 5% Standard Deviation (SD) from medical records dataset. We identified the age-group and gender of the patient, and the RF model outperformed others with the best accuracy 96% and 94% in both cases having 11% SD. Also, prominent features such as CPK, SC, follow-up month, platelets, and ejection were found to be significant factors in predicting the patient&rsquo;s age-group. Smoking habits, CPK, platelets, follow-up month, and SC of each patient were discovered to be significant predictors of patient gender. The hypothetical study proved that SC and SS making substantial differences in the survival of patients (p &amp;lt; 0.05) and failed to reject that anemia, diabetes, and BP making a significant impact on the creatinine and sodium of each patient (p &amp;gt; 0.05). With &chi;2(1) = 8.565, the Kaplan Meier plot revealed that mortality was high in the extremely elder age-group. The finding has possible effects on clinical practice and becomes a new medical support system when predicting whether a patient can survive a heart attack or not. The doctor should primarily concentrate on follow-up month, SC and EF, CPK, and platelet count since the aim is to understand whether a patient survives after HF.

scholarly journals Level of Combined Estrogen and Progesterone Receptor Expression Determines the Eligibility for Adjuvant Endocrine Therapy in Breast Cancer Patients

Cancers ◽  
2021 ◽  
Vol 13 (19) ◽  
pp. 5007
Author(s):  
Jee Hyun Ahn ◽  
Soon Bo Choi ◽  
Jung Min Park ◽  
Jee Ye Kim ◽  
Hyung Seok Park ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Progesterone Receptor ◽  
Endocrine Therapy ◽  
Adjuvant Endocrine Therapy ◽  
Receptor Expression ◽  
Breast Cancers ◽  
Breast Cancer Patients ◽  
Limited Data ◽  
Single Receptor

Hormone receptor (HR)-positive breast cancer has a heterogeneous pattern according to the level of receptor expression. Patients whose breast cancers express low levels of estrogen receptor (ER) or progesterone receptor (PgR) may be eligible for adjuvant endocrine therapy, but limited data are available to support this notion. We aimed to determine whether HR expression level is related to prognosis. Tumors from 6042 patients with breast cancer were retrospectively analyzed for combined HR levels of ER and PgR. Low expression was defined as ER 1–10% and PgR 1–20%. Four HR groups were identified by combining ER and PgR expression levels. Patients whose tumors expressed high levels of a single receptor showed the worst survival outcomes, and their risk continuously increased even after the 10-year follow-up. Endocrine therapy had a significant benefit for patients whose tumors expressed high HR levels and a favorable tendency for patients with tumors expressing low HR levels. We established the possible benefit of endocrine therapy for patients whose breast tumors expressed low HR levels. Thus, HR level was a prognostic factor and might be a determinant of extended therapy, especially for patients with high expression of a single receptor.

Abstract PD2-07: 10-year follow-up and biomarker discovery for adjuvant endocrine therapy; results of the TEAM trial

2017 ◽  
Author(s):  
EJ Blok ◽  
MGM Derks ◽  
PJK Kuppen ◽  
EM Meershoek-Klein Kranenbarg ◽  
CC Engels ◽  
...  
Keyword(s):  
Endocrine Therapy ◽  
Biomarker Discovery ◽  
Adjuvant Endocrine Therapy
Sign in / Sign up

Export Citation Format

Share Document