Abstract
Abstract
Background. Familial clustering of PV, ET, MF, and chronic myeloid leukemia (CML) has been reported through case reports and smaller case series. Recently, several studies reported that in families with multiple MPD patients, the JAK2 mutation is not an early germ line predisposing factor for MPD but rather a facilitator of proliferative advantages. However, the degree of familial clustering in the population has not been defined. We have conducted the first large population-based study to quantify risks of developing MPD and related malignancies in first-degree relatives of MPD patients.
Methods. Using high-quality central population-based registries, we identified 6217 PV, 2838 ET, and 1984 MF patients diagnosed in Swedish hospitals 1958–2005 (Cancer and local hospital-based registries), with linkable relatives; 43550 frequency-matched controls (Population registry); and first-degree relatives of cases (n=24577) and controls (n=99542) (Multigenerational registry). Relatives of MPD patients and controls were linked with the Cancer and local hospital-based registries to define occurrence of MPD and other related neoplasms. We used a marginal survival model to calculate relative risks (RR) and 95% confidence intervals (CI) as measures of familial aggregation.
Results. Compared with controls, relatives of MPD patients had significantly increased risks of PV (RR=5.70; 95% CI 3.55–9.14), ET (RR=7.37; 95% CI 3.67–14.81), and MF (RR=3.53; 95% CI 1.59–7.85). Risks were similar when we restricted the analyses to relatives of patients with the same MPD (PV-PV, ET-ET, and MF-MF) Also, risk-estimates were virtually the same for various types of first-degree relatives (parents, siblings, offspring); the same was true when we calculated risks by age at MPD of cases (above vs. below 65 yrs), and sex of relatives. Age at diagnosis of MPD was not different for case and control relatives. Furthermore, relatives of MPD patients (vs. controls) had a borderline increased risk of CML (RR=1.86; 95% CI 0.90–3.74; p=0.07). As expected, we found excess of subsequent acute myeloid leukemia (AML) (n=271; 2.5%) and myelodysplastic syndrome (MDS) (n=27; 0.2%) among MPD patients; however, there were no increased risks of AML or MDS among relatives of MPD patients.
Conclusions. In this first large population-based study including more than 11000 MPD patients and their almost 25000 linkable first-degree relatives, we found 3- to 7-fold elevated risks of developing MPDs among first-degree relatives of MPD patients. Our results support the hypothesis that there are common, strong, shared susceptibility genes that predispose to PV, ET, MF, and possibly CML.
