2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Access, Representation, Population Health, Technology and Procedures, Imaging, Treatment Considerations, Supportive Care, Biological Therapies, Transplantation, Bispecific Antibody Therapy, (CAR)-T Cell Therapies, Workforce, Adverse Events, Artificial Intelligence (AI), Generative AI, Natural Language Processing, Biological Processes, Pathogenesis, Metabolism, Clinical Practice (Health Services and Quality), Clinical Guidelines, Health Economics, Patient Safety, Practice Models, Quality Improvement, Quality of Care, Value, Diseases, Immunology, Immunodeficiency, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, CLL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Indolent Lymphoma, Aggressive Lymphoma, Research, Translational Research, CML, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Health Disparities Research, Patient-Reported Outcomes, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult, Elderly
PAPER

4409 What best predicts clinical trial participation in older adults with hematologic malignancies: Frailty, multimorbidity, or polypharmacy?

Sudipto Mukherjee, MD, PhD, MPH, Fatima Hussain, MPH2, Vu Long, MS3, Christopher Swetel, N/A4, John Carlos Molina, MD, MEd5, Hetty Carraway, MD, MBA6, Akriti G Jain, MD, FACP7, Aaron T. Gerds, MD, MS8, Deepa Jagadeesh, MD, MPH9, Johnie Rose, MD, PhD10, Jack Khouri, MD11, Brian Hill, MD, PhD12, Sophia Rebecca Balderman, MD13, Moaath Mustafa Ali, MD, MPH14, Anjali Sunder Advani, MD15, Abhay Singh, MD, MPH16, Faiz Anwer, MD17, Paolo Fabrizio Caimi, MD18, Siran M. Koroukian19
1 Cleveland Clinic, Cleveland, Ohio, United States
2 Case Western Reserve University, Cleveland, Ohio, United States
3 Case Western Reserve University, Cleveland, Ohio, United States
4 Case Western Reserve University, Cleveland, Ohio, United States
5 Cleveland Clinic, Cleveland, Ohio, United States
6 Cleveland Clinic, Cleveland, Ohio, United States
7 Cleveland Clinic, Cleveland, Ohio, United States
8 Cleveland Clinic, Cleveland, Ohio, United States
9 Cleveland Clinic, Cleveland, Ohio, United States
10 Case Western Reserve University, Cleveland, Ohio, United States
11 Cleveland Clinic, Cleveland, Ohio, United States
12 Cleveland Clinic, Cleveland, Ohio, United States
13 Cleveland Clinic, Cleveland, Ohio, United States
14 Cleveland Clinic, Cleveland, Ohio, United States
15 Cleveland Clinic, Cleveland, Ohio, United States
16 Cleveland Clinic, Cleveland, Ohio, United States
17 Cleveland Clinic, Cleveland, Ohio, United States
18 Cleveland Clinic, Cleveland, Ohio, United States
19 Case Western Reserve University, Cleveland, Ohio, United States

Background
Clinical complexity, characterized by the presence of multiple chronic conditions (MCCs), polypharmacy, and/or frailty, has been identified as important factors in clinical trial participation (CTP).  Indeed, those with MCCs, polypharmacy, or frailty are less likely than their healthier counterparts to participate in trials. We sought to determine the discriminative ability of each of these measures in predicting CTP in older adults with hematologic malignancies (HMs).

Methods
We used SEER-Medicare data for older adults newly diagnosed with HMs during the years 2014-2021 and receiving their care through the fee-for-service system. Starting in 2014, the Centers for Medicare & Medicaid Services required that the National Clinical Trial number, which, when linked with ClinicalTrials.gov data, it makes it possible to identify individuals enrolled in trials targeted to patients with HMs. We used the SEER database to retrieve sociodemographic variables (age, sex, race/ethnicity, and marital status). From the Medicare Beneficiary Summary File, we flagged 67 chronic and potentially disabling conditions identified by CMS from claims data and grouped them into 14 clinically meaningful categories. From Part D data, we identified individuals with polypharmacy, categorized as < 5, 5-8, or 9+ medications. Lastly, we used Kim’s algorithm to derive frailty measures from claims data, categorized as non-frail, pre-frail, and frail. In addition to descriptive analysis, we created several logistic regression models with CTP as the outcome. Independent variables in all models included demographics (age, sex, race/ethnicity), cancer types, residence in Metro and non-Metro counties. The first model included the 30 chronic conditions; the second model included all 30 chronic conditions and the 37 potentially disabling conditions (e.g., mobility impairment, psychiatric conditions, and pain-related conditions); the third model included the 14 clinically meaningful categories (by organ systems); and the fourth set of models included frailty. Each of these models was tested before and after including polypharmacy. Models including polypharmacy were limited to those enrolled in Medicare Part D, representing 63% of the population. We compared the models’ discriminative ability using the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Nageikelke R2 (or pseudo R2) measures.

Results
Our study population included 169,514 older adults with incident HMs, of whom 2.9% participated in clinical trials. The median age was 76 years (IQR: 71-83); 44.1% were women; and 83.1% were non-Hispanic White. Lymphomas accounted for the largest diagnostic group (43.5%), followed by lymphoid leukemias (20.2%), multiple myeloma (19.7%), and myeloid leukemias (13.1%). With respect to frailty, 38.0% of patients were non-frail, whereas 44.4% were pre-frail and 17.9% were frail. Polypharmacy was common: only 16.3% of patients used fewer than 5 medications, while 26.7% used 5-8 medications and 57.0% used 9 or more medications. Models based on the 30 chronic conditions, the full set of 67 chronic and potentially disabling conditions, or the 14 clinically meaningful categories demonstrated similar discrimination, with AIC values ranging from 26946 to 26961, BIC values from 27137 to 27152, and Nageikelke R2 values from 0.116 to 0.118, regardless of whether polypharmacy was included. In contrast, models based on frailty performed modestly better, with AIC and BIC values decreasing to 26737 and 26929, respectively, and Nageikelke R2 increasing to 0.128. These results indicate that frailty provided slightly better discrimination for CTP than multimorbidity or polypharmacy.

Conclusions
In this first analysis evaluating up to 67 chronic and potentially disabling conditions, polypharmacy, and frailty, CTP models showed similar discrimination across condition-based measures and with or without polypharmacy. Frailty modestly outperformed multimorbidity and polypharmacy, suggesting that functional vulnerability captures CTP variation better than disease or medication burden alone. As a claims-based proxy for functional status measures such as ECOG or Karnofsky scores, frailty should be prioritized when designing strategies to improve representation of frail older adults in hematologic malignancy trials.

Sudipto Mukherjee, MD, PhD, MPH: Sobi - Consultancy (Includes expert testimony); Recordati - Consultancy (Includes expert testimony); CCM Biosciences - Consultancy (Includes expert testimony); Jazz Pharmaceuticals - Consultancy (Includes expert testimony); McGraw Hill - Consultancy (Includes expert testimony); Blueprint Medicines - Consultancy (Includes expert testimony)