Background and Objective: Hairy cell leukemia (HCL) is a rare condition and accounts for ~2% of all leukemia cases in the US. NCCN guidelines recommend first-line agents including pentostatin and cladribine among patients with HCL. However, a paucity of data exists with regard to real-world treatment patterns among patients with HCL. Current study evaluates treatment patterns and associated clinical outcomes among patients with HCL using a large US administrative claims database.
Methods: This retrospective observational study was conducted using the IMS Health PharMetrics Plus Health Plan Claims Database (2006-2014), which includes over 150 million unique health plan members across the US and is nationally representative of the commercially-insured US population. Data includes date stamped medical and pharmacy records along with information on health plan enrollment. Individuals with at least 2 medical claims with a diagnosis for HCL (identified using ICD-9-CM cod: 202.4x) were selected and the first observed claim defined the "index date." Patients <18 years of age at index date or with other malignancies during 6 months (the "pre-index period") prior to index date were excluded. Patients were required at least 90 days of continuous enrollment (the variable "follow-up period") in the health plan post index date with exception of patients who died within 3 months of diagnosis. Patients were followed until death (recorded on inpatient discharge disposition), end of enrollment or end of database, whichever occurred earlier. Study measures including patient demographic and baseline clinical characteristics, line of therapy (LOT), treatment patterns, relapse (receipt of same or new regimen in subsequent LOT following a gap of 365 days) and refractory disease (receipt of same or new regimen where the gap of two adjacent LOTs was <365 days) and post-treatment complications were assessed during the follow-up period. All analyses were descriptive in nature.
Results: The study cohort included 749 patients after applying the selection criteria (mean follow-up from diagnosis 32 months). At diagnosis, the mean age (standard deviation) of the study cohort was 56 (10) years and majority of patients were male (77%). Mean baseline comorbidity burden (assessed using Charlson Comorbidity Index score) was 0.8 (1.1) with hypertension (24%) and aplastic anemia (22%) being the two most common co-morbidities. Only 38% (n=282) of patients received first-line chemotherapy post diagnosis. Majority initiated first-line cladribine (76%) as a single agent, while 9% had evidence of single agent pentostatin. Mean time to initiation of first-line therapy from diagnosis was 132 (294) days and average time on first-line therapy was 34 (104) days. Among patients with first-line therapy 14% received second-line therapy and rituximab (53%) and cladribine (21%) were frequently observed second-line agents. Post first-line therapy, mean time to initiation of second-line therapy was 303 (406) days. Among second-line initiators, 76% had refractory disease and 24% had relapsed following first-line. Neutropenia and fever were frequently reported complications while on chemotherapy.
Conclusion: The real-world chemotherapy utilization patterns observed in this study are consistent with the NCCN guidelines with cladribine and pentostatin being the agents of choice for first-line therapy. Following diagnosis, more than one-third of patients initiated chemotherapy and only a small proportion of these received second-line chemotherapy suggesting durable response with first-line therapy. Limited follow-up post first-line therapy may have impacted the proportion of patients initiating second-line therapy as well as categorization of refractory and relapse disease.
Karve:AstraZeneca: Employment. Divino:IMS Health: Employment, Other: IMS Health received funding from AstraZeneca for this study. Gaughan:AstraZeneca: Employment. DeKoven:IMS Health: Employment, Other: IMS Health received funding from AstraZeneca for this study. Gao:MedImmune: Employment. Lanasa:MedImmune: Employment.
Asterisk with author names denotes non-ASH members.