Cancer Risk in Patients with Gaucher Disease Using Real-World Data

Abstract

Gaucher disease (GD), a rare autosomal recessive disorder, is characterized by the deficiency of the lysosomal enzyme β-glucocerebrosidase, leading to the accumulation of glucosylceramide in macrophages throughout the body. Accordingly, the primary disease features are hepatosplenomegaly, hypersplenism, anemia, thrombocytopenia, and bone involvement. There are three types of GD, with a combined estimated occurrence ranging from 0.45 to 25.0 per 100,000 live births. However, among individuals with Ashkenazi Jewish heritage, type 1 GD is considerably more prevalent, with an estimated occurrence of 1 in 850 live births. Considerable controversy exists regarding the co-occurrence and risk of other diseases in patients with GD compared to the general population, GD-related and GD-unrelated, respectively. While Parkinson disease and cholelithiasis are clear examples of GD-related co-morbidities, there is uncertainty regarding the association between GD and cancer, a topic that raises great concern for patients and their families. The aim of the current study was to evaluate the risk of cancer in patients pre- and post- diagnosis of GD compared to a group of matched controls in a population-based comprehensive clinical database. This study utilized electronic health records from the Maccabi Healthcare Service (MHS), Israel's second-largest health maintenance organization, to study cancer risk in 264 patients with GD compared to 3441 controls matched by sex, year of birth, and socioeconomic status (SES). Cancer diagnoses were extracted from the MHS cancer register, which captures all cancer types, updated until January 31, 2021. Additional data extracted included ICD-9 codes for family history of cancer, smoking, splenectomy, the use of Gaucher-specific medication, and MHS codes for colonoscopy, mammography, occult blood in stool, Papanicolaou (PAP) smear, prostate-specific antigen (PSA), and skin screen. We analyzed cancer occurrence before and after the index date for GD diagnosis (i.e., the first documentation of GD in the MHS database). For the controls, the index date used was the index date for the GD diagnosis of the matched patient. The study design received approval from the MHS Institutional Review Board. Patient consent was waived since aggregated, de-identified, and anonymized data were utilized Cancer was diagnosed in 18 individuals before the index date; 11 (4.2%) in GD compared to 7 (0.2%) in controls (p< 0.001). The main types of cancers in patients with GD were non-melanoma skin and hematological cancers, with clustering of the diagnoses around the time of GD diagnosis. Cancer was diagnosed in 57 individuals (20 (7.9%) in GD and 37 (1.1%) in control) after the index data, with a median follow-up of almost 13 years (Table). Patients with GD had a significantly higher risk of developing cancer than controls (Figure). The age and sex-adjusted incidence of cancer (excluding non-melanoma skin cancer) was 4.1 (95% CI 1.9-6.4) per 1,000 patients-years in patients with GD compared to 0.65 (95% CI 0.4-0.9) per 1,000 patients-years in controls, with an incidence rate ratio of 6.4. The median (range) age at cancer diagnosis was 62 (36-79) and 46 (26.5-91) years in patients with GD and controls, respectively (p=0.02). Multiple primary tumors were reported in four patients with GD but none in controls. Patients with GD underwent more cancer screening tests, e.g., colonoscopy, prostate-specific antigen, and skin screening, compared to controls. No difference in the rate of family history of cancer or smoking between patients with GD and controls. In a multivariate analysis among patients with GD, including age, sex, SES, the use of Gaucher-specific medication, and splenectomy, age was the only risk factor for cancer. The primary significance of this matched case-control study is that it addresses the cohort/selection bias present in previous reports regarding the potentially heightened cancer risk among individuals with GD, where compression of rates was done with non-matched cohorts. We herein show that the increased occurrence of cancer in patients with GD can be attributed partly to a more rigorous surveillance approach and frequent screening. However, the higher rates of cancers without early detection measures, such as multiple myeloma, indicate that there is indeed an elevated risk of developing cancer in individuals with GD that is not related to an ascertainment bias.