Real-world (RW) testing patterns of serum lactate dehydrogenase (LDH) and its prognostic value among patients (pts) with metastatic colorectal cancer (mCRC).

Abstract

46 Background: LDH is a potential prognostic biomarker of outcomes for pts with mCRC. This retrospective, observational cohort study describes RW LDH testing patterns and assesses LDH as a prognostic factor for overall survival (OS) in pts receiving chemotherapy (CT) for mCRC in the USA. Methods: Pts with mCRC, ≥18 years (yrs) of age, who initiated first line (1L) CT (index date) between January 1, 2016, and November 30, 2022, were identified in the nationwide de-identified Flatiron Electronic Health Record-derived database. Practice patterns for LDH testing are described. Pts were categorized into normal vs abnormal LDH baseline (≤3 months prior to/at index date) value based on the lab reference range. Prognostic relationship between normal vs abnormal LDH baseline value and OS was assessed using Kaplan-Meier and multivariate Cox proportional-hazards model. Results: Of 15,329 eligible pts, 57% were male, 49% were ≥65 yrs of age and 66% had ECOG performance status 0–1; 59% had stage IV disease at initial diagnosis; 29% had a KRAS mutation (at any time ≤30 days post-index; 35% not tested). Median time from mCRC diagnosis to index was 35 days. Most common 1L regimens were FOLFOX ± bevacizumab (48%), FOLFIRI ± bevacizumab (15%), and capecitabine (13%). Median follow up from index date was 15.6 months (mos); 60% died. LDH testing was performed in 2,547 pts (17%) during baseline and 3,379 pts (22%) post-index. Among those with ≥1 LDH test during each line of therapy (LOT), median number of LDH tests per pt was 4–6 and median time from line initiation to first LDH test was 7–10 days across the first five LOTs (Table). Among LDH tested pts during follow-up (n=3,379), 21% and 40% had abnormal and normal LDH values at baseline, respectively (39% missing). Demographic characteristics of abnormal vs normal LDH pts were similar; abnormal LDH pts were more likely to be ≥65 years (52% vs 49%), African American or Black (12% vs 7%), and reside in the Northeast (30% vs 21%). Median OS (95% CI) in pts with normal vs abnormal baseline LDH was 29.9 (28.3, 31.7) mos vs 16.8 (14.8, 18.2) mos. In multivariate Cox proportional-hazards model, pts with abnormal baseline LDH values had higher risk of death (HR 1.91, p<0.0001) after adjustment for pt demographic and clinical characteristics. Conclusions: In RW clinical practice in the USA, only one-fifth of mCRC pts receiving CT undergo LDH testing even though LDH value has been found to be a prognostic factor of OS. Efforts to increase LDH testing may help guide treatment decisions in this setting. [Table: see text]