Evaluating the Carg Chemotherapy Toxicity Calculator Among Older Adults Newly Diagnosed with Hematologic Malignancy

Background: Older adults with hematologic malignancy (HM) are a growing demographic and providing effective treatments that balance toxicity and health related quality-of-life (HRQL) is imperative. The well-studied and utilized chemotherapy toxicity tool, the Cancer and Aging Research Group (CARG) chemotherapy toxicity score, has not been validated in hematologic malignancies.

Methods: The primary objective of this study was to validate the predictive ability of the CARG score for grade 3-5 toxicity in newly diagnosed (ND) patients >60 years with HM. This was a prospective longitudinal study with 4 study visits: baseline (pre-therapy), visit 2 (Day 90), visit 3 (Day 180), and end-of-study (EOS) which occurred at the earliest of the following events: progression, transplant, or 1-year from baseline. The CARG score was evaluated at baseline. HRQL (PROMIS-GHS) and physical function measured by short physical performance battery (SPPB) were assessed longitudinally at all visits. Treatment toxicity using the NCI CTCAE (version 5.0) were captured monthly, and the worst grade of each type of chemo-related adverse event (AE) was recorded and summarized for each patient. Wilcoxon signed-rank test was used to test if variables changed significantly across visits. Fine and Gray model was used to associate comprehensive geriatric metrics and CARG score with the development of grade 3-5 chemotherapy-related toxicity with death as the competing risk, and Cox model was used to analyze overall survival (OS).

Results: Ninety-seven patients with hematologic malignancy (myeloid n=34, lymphoma n=35, plasma cell n=28) were enrolled. The median age was 70 years (range 60-88) with a median 159 days on study (range 1-435). Baseline evaluations: ECOG PS was 0-1 in 69 (85%), median IADL score was 13 (range 5-14), median MOS physical health score was 44.4 (range 0-100), median self-reported KPS was 80% (50-100%), and median comorbidity score was 6 (range 2-12). PROMIS median scores improved from baseline (32, range: 12-49) to EOS (35, range: 16-47, p=0.05). Median SPPB scores improved significantly from baseline (5, range 0-12) to EOS (9, range 0-12, p=0.005). During the study period, 75 patients had 334 grade 1-2 AEs, and 42 patients had 82 grade 3-5 AEs. Hematologic toxicities were more common with 36 (37%) patients having anemia (30 grade 1-2, and 6 grade 3-5) and 11 (11%) patients having febrile neutropenia (all grade 3-5). In multivariable analysis, significant risk factors associated with grade 3-5 toxicity (p<.05) included living alone (HR 3.11, 95%CI: 1.52-6.34) and social activities score (HR 1.21, 95%CI: 1.02-1.42). Risk factors associated with OS in univariable models (p <0.05) were ECOG PS (HR 4.35, 95%CI 2.43-7.79), physical health score (HR 0.86, 95%CI 0.77-0.97), IADL (HR 0.83, 95%CI 0.73-0.95), comorbidities (HR 1.26, 95%CI 1.07-1.48), number of supplements (HR = 0.76, 95% CI 0.60-0.96), memory score (HR 1.08, 95% CI 1.01-1.14), SPPB score (HR 0.87, 95%CI 0.80-0.95) and CARG score (HR 1.12, 95%CI 1.02-1.23). In multivariable analysis, the SPPB score (HR 0.85, 95%CI 0.78-0.93) remained significant for OS.

Conclusions: The CARG chemotoxicity score was not predictive of grade 3-5 toxicities in patients ND with HM, but was univariably associated with OS. Higher SPPB scores were strongly associated with OS. Future studies, evaluating modifications of the CARG score are indicated for patients with HM. Objective measures of function, such as the SPPB, may be a reliable method to stratify treatment intensities for older adults with HM.