Mitochondrial Haplotype As an Independent Predictor of Outcomes Following Allogeneic Hematopoietic Cell Transplantation (HCT)

Mitochondria (mt) provide cells energy through oxidative phosphorylation (OXPHOS), regulate cell survival and death, and likely affect innate and adaptive immune system responses. Polymorphisms in mtDNA can be grouped into haplotypes (mthaps) that are associated with human global migration. Experiments using cybrids (identical cell clones that differ in mthaps) show differences in OXPHOS, reactive oxygen species generation and immune recognition. Numerous association studies have linked specific mthaps to disease occurrence, severity and/or therapy response. Since HCT is a period of high metabolic demand, we explored whether patient or donor mthaps are associated with HCT outcomes. Pre-transplant DNA was available from 437 adult and pediatric patients (6 months-69.6 years) who received an allogeneic transplant at the University of Minnesota for a hematological malignancy between 1995 and 2005, along with DNA from 327 donors. Of the HCTs, 213 were related donor (198 siblings, 15 other), 73 unrelated and 151 umbilical cord blood. Clinical and laboratory data were collected on all patients. Patient and donor DNA were genotyped for 11 common European mthaps (most to least common: H, J, U, T, Z, K, V, X, I, W, K2); 29 samples did not correspond with one of the 11 mtHaps and were labeled as ‘other’. Three separate analyses were performed: 1) sibling transplant outcomes (matched for mthap), 2) related and unrelated transplant outcomes based on patient mthap, and 3) related and unrelated transplant outcomes based on donor mthap. Multivariate models (adjusted for stem cell source, disease risk, conditioning regimen, patient CMV serostatus and patient sex as appropriate) were used to calculate relative risks (RR) and 95% confidence intervals (CI) comparing HCT outcomes to the most common haplotype (H). More than 80% of the 11 mthaps occurred in non-Hispanic whites, although 47% of Z (7/15 patients) were of other race/ethnicity. Among matched siblings, mthap did not differ with HLA mismatch (p=0.59). Compared to mthap H (43 events/83 patients, 52%), recipient siblings with I (5/6) and V (5/5) were significantly more likely to die within 5 years (RR=3.0, 95%CI 1.2-7.9, p=0.02; 4.6, 1.8-12.3, p<0.01, respectively). Patients with W (3/4) experienced significantly higher aGVHD II-IV events (RR=2.1; 1.1-2.4, p=0.03) than H (32/83, 39%), while K2 (0/4), K (1/8) and J (4/21) experienced no or fewer events. For aGVHD III-IV, patients with U (5/19) and V (2/5) experienced significantly higher events than H (9/83, 11%); there were no events for K and K2. Similar results were observed when considering outcomes for all recipients: compared to H (78/156, 57%), fewer aGVHD II-IV events were observed for J (17/59) and K (6/22) patients (0.5, 0.3-0.8, p<0.01; 0.4, 0.2-0.9, p=0.03, respectively). Lastly, when comparing donors relative to H (68/127, 54%), 6/7 recipients with I donors died within 5 years post HCT (2.7, 1.2-6.2, p=0.02), while only 1/7 and 1/4 patients receiving a transplant from a K2 or W donor, respectively, died. Further, compared to H donors (27/127, 23%), no patients who received a HCT from a K2 donor relapsed and only 4/34 patients from U donors relapsed. Ours is the first study to demonstrate that mthaps may be an important consideration in determining patient outcomes following HCT. Given interest in mthaps in disease susceptibility, we will conduct functional studies to better understand the role of patient and donor mthaps in OXPHOS and immune system responses in the presence of HCT. These preliminary results need confirmation in larger studies. If validated, it would be feasible to select donor mthaps associated with less GVHD and/or relapse (e.g., J, K2, U) or avoid use of donor mthaps associated with adverse outcomes (e.g., I,V).