No familial aggregation in chronic myeloid leukemia

To the editor:

There is a fivefold to sevenfold elevated risk of myeloproliferative neoplasms (MPNs) among first-degree relatives of MPN patients.^1,2  In contrast, aside from early-onset patients, there is no significant familial aggregation in acute myeloid leukemia (AML).³ 

The molecular underpinnings of the development of chronic myeloid leukemia (CML) are unclear. Ionizing radiation in high doses is the only known risk factor.⁴  Benzene and benzene-containing products have been reported to be significantly related to morbidity and mortality from CML,⁵  although recent case-control literature indicates the opposite.⁶  Apart from extremely rare pedigrees with multiple cases of CML/myeloproliferative disorders,⁷  there is essentially no data on familial aggregation of CML in the population. According to The National CML Society, “Occasionally, there are families that may have other members living with leukemia, however, there is no conclusive evidence that family members are predisposed to develop leukemia.”⁸ 

Taking advantage of high-quality registry data from Sweden, we conducted a population-based registry study to evaluate risk of CML, MPN, AML, and other malignancies among 9491 first-degree relatives of 4619 patients (45% females) with CML compared with 42 474 first-degree relatives of matched controls. Our methods have been previously described.³  In brief, using the Swedish Cancer Registry, we identified all patients with a primary diagnosis of CML diagnosed between 1958 and 2004. For each CML patient, 4 population-based controls (matched by sex, year of birth, and county of residence) were chosen randomly from the Swedish population database. All control individuals had to be alive at the time of CML diagnosis for their corresponding case patient and without a hematologic malignancy at the date of CML diagnosis for their corresponding case patient. We obtained information on all first-degree relatives of cases and controls from the Swedish Multigenerational Registry, which includes data on parent-sibling-offspring relations for all Swedish citizens born in 1932 or later. We used a marginal survival model to calculate familial aggregation.

We found that neither CML (relative risk, 1.62; 95% confidence interval, 0.52-5.11), AML (0.94; 0.44-2.0), nor MPN (1.11; 0.58-2.17) aggregated significantly in relatives of patients with CML (vs relatives of controls) (Table 1). In addition, the relative risks for any lymphoproliferative, hematologic, or solid tumor were not significantly increased. We also analyzed the risks in relatives by gender and age at diagnosis of proband (≤60 vs >60 years) and could not reveal any significant associations (data not shown).

CML patients may worry about their family members having a potentially increased risk of developing CML or other cancer. In this population-based study, including all age groups of patients with CML, we found no significant familial aggregation for CML, AML, MPN, lymphoproliferative neoplasm, or any other cancer among relatives of CML patients (vs relatives of matched controls). Our findings are in sharp contrast to those observed in MPNs where an increased familial aggregation supports the notion that genetic susceptibility genes may play a strong role in these patients.