Patients with Fanconi Anemia Have an Increased Incidence of Diabetes Mellitus After Allogeneic Stem Cell Transplantation

Fanconi anemia (FA) cells are characterized by genomic instability, which places FA patients at risk for malignancies and multiple other functional long-term abnormalities. Hematopoietic cell transplantation (HCT) restores normal hematopoiesis in these patients.1–3 Endocrinological abnormalities are frequent in FA patients; diabetes mellitus (DM) has been shown to occur at a higher frequency in FA patients both pre- and post-HCT.4–8 HCT recipients in general, however, have also been shown to be at an increased risk of DM.9 The data on the incidence of DM in FA patients post-HCT are scarce, so we designed this retrospective to study the impact of HCT on DM development in FA patients and compare its incidence to that in a control group of patients who received HCT for non-FA acquired aplastic anemia (SAA). The total cohort consists of 320 pediatric patients who underwent 351 allogeneic (HCT) at our institution from 1 January 1993 to December 2020. Of those, 127 (39.7%) underwent HCT for SAA and 193 (60.3%) underwent HCT for FA. Median age at transplant was 8.4 years (range, 0.9–17.4 years). The Severe Aplastic Anemia Subgroup A total of 127 patients underwent 143 transplants in this subgroup; 114 had single transplants. Median age at HCT was 8.7 years (range, 1.2–14.6 years). Radiation (XRT) was a part of conditioning in 21 patients (14.7%). The source of stem cells was bone marrow (BM) in 120 (83.9%), peripheral stem cell collection (PBSC) in 8 (5.6%), and cord blood (CB) in 15 (10.5%) patients. The donor was HLA identical in 104 (72.7%), Haploidentical in 17 (11.9%), related 1-AG mismatched in 7 (4.9%), and matched unrelated CB in 15. ANC engraftment rate was 88.8% (n = 127) with a median time to ANC recovery of 17 days (range, 10–49 days). Cumulative incidence of all-grade acute GvHD was 18.9% (n = 27). Post-HCT incidence of DM in this subgroup was 0.8% (n = 1 of 127); the patient was insulin-dependent and was alive at the last follow-up. Mortality rate was 23.6% (n = 30), a median follow-up time was 104.8 months (95% CI: 85.6–124.1), and the cumulative probability of 5-year overall survival (OS) was 78.5% ± 3.7%. The FA subgroup In total, 193 patients underwent 208 transplants in this subgroup; 179 had single transplants. Median age at transplant was 8.3 years (range, 0.9–17.4 years). XRT was a part of conditioning in 55 (26.4%). Source of stem cells was BM in 182 (87.5%), PBSC in 8 (3.8%), BM+PBSC in 2 (1.0%), and CB in 16 (7.7%). The donor was HLA identical in 142 (68.3%), Haploidentical in 39 (18.8%), related 1-AG mismatched in 9 (4.3%), related 2-AG mismatched in 2 (1.0%), and matched unrelated CB in 16 patients. ANC engraftment rate was 92.0% (n = 185) with a median time to ANC recovery of 14 days (range, 9–50 days). Cumulative incidence of all-grade acute GvHD was 23.6% (n = 49). Post-HCT incidence of DM in this subgroup was 4.1% (n = 8 of 193); two of them (25.0%) had received XRT in their conditioning (P = 1.00). Five (62.5%) of those with DM were insulin-dependent, and 3 (37.5%) were on metformin. With a mortality rate of 22.3% (n = 43) and a median follow-up time of 101.7 months (95% confidence interval: 80.6–122.9), the cumulative probability of 5-year OS was 84.8% ± 2.6%. No significance difference was found for OS between DM (+) and DM (−) cases (51.4%±20.4 vs. 52.1±11.7% (P = 0.509), Tarone–Ware test. Overall, the incidence of DM was 2.8% (n = 9 of 320 cases). We found a trend of a higher incidence of DM in FA patients (4.1%) when compared to SAA patients (0.8%), although it did not reach statistical significance (P = 0.093). Allogeneic HCT has dramatically improved the outlook for FA patients as it restores normal hematopoiesis in these patients with favorable success rates, using matched related, matched unrelated, and haploidentical HCT.1–3 But despite the encouraging results, more is yet to be documented on the long-term effects of the conditioning regimens on the nonhematopoietic FA cells in HCT recipients. FA patients remain subject to multiple long-term complications after HCT, such as secondary cancers and endocrinological problems. Many of the FA patients have at least one endocrine abnormality such as short stature, GH deficiency, abnormal glucose or insulin metabolism, hypothyroidism, hypogonadism, or impaired fertility. In our institution, we recently published our data and showed that 36 out of 87 surviving FA patients had evidence of endocrinopathy/gonadal dysfunction.10 Diabetes and/or impaired glucose metabolism represent one of the major endocrine problems to occur in FA patients, with a prevalence of diabetes is approximately 8%–10%, whereas an additional 27%–68% may have impaired glucose tolerance.4–8. This problem has been reported to occur pre- and post-HCT; the exact incidence post-HCT is not however clear in the literature. Giri et al. reported abnormal glucose homeostasis (insulin resistance, hyperglycemia, glucose intolerance, and DM in 39% of 45 patients (nine of the patients had undergone HCT).4 On the other hand, Polgreen et al. reported in a cross-sectional study of glucose and insulin metabolism in 17 children with FA HCT that many of them had normal glucose tolerance and normal beta-cell function after HCT.11 In a study by Wajnrajch et al., 25% of the children with FA had impaired glucose tolerance by oral glucose tolerance testing (OGTT), but the majority of these patients (94%) had not had HCT.6 In addition, a recent study by Elder et al. found that 46% of children with FA before HCT had impaired glucose tolerance (IGT) or DM, also suggesting an innate defect in FA.7 Increased incidence of DM, on the other hand, has also been reported in pediatric HCT survivors in general. In a study by Taskinen et al., 12 (52%) of 23 post-HCT pediatric patients had insulin resistance, including four patients with DM.9 In a larger cohort of 748 patients reported by Hoffmeister et al., with a follow-up median of 11 years, the prevalence of type I diabetes among surviving patients was three times higher than the normal population (0.52%), and that of type II diabetes was higher than expected in leukemia survivors (9%) and among aplastic anemia survivors (2%).12 In this study, we found that patients with FA post-HCT have a 4.1% incidence of DM, higher than that of the comparison group (0.8%). As pointed out earlier, this could merely reflect the natural history of the FA cells; however, an added effect of HCT cannot be ruled out. In our study and during the specified follow-up time, DM did not affect survival, but longer follow-up is crucial to study the effect of the development of DM on FA survivors.