Background/Purpose: Bone marrow lesions (BMLs) have been consistently associated with the development of osteoarthritis (OA); however the extent to which they are genetically driven by mitochondrial genetic background is unknown. Since specific mtDNA haplogroups have been associated with decreased risk of incident knee OA, the aim of this work was to examine the associations between mtDNA haplogroups and MRI-detected BMLs

Methods: Knees that did not develop either incident OA or radiographic progression during 48 months follow-up were identified from 1030 Osteoarthritis Initiative (OAI) subjects. The most common Caucasian mtDNA haplogroups (H, Uk, J, T and Others) were assigned to all participants using the single base extension (SBE) assay. BML measurements from MR images acquired using IW-TSE sequence with sagittal slices in different knee regions were obtained annually between baseline and 24 months using an automated system, and were expressed as % of the bone region. Logistic regression with generalized estimating equations and multinomial logistic regression models were used to evaluate the association between mtDNA haplogroups and i) large BMLs, present at any point between baseline and 24 months and defined as ≥33% of the subregional bone volume, as well as ii) the progression of BMLs at 24 months according to an increase of ±1% volume between baseline and final visit. Both models were adjusted by confounder variables of gender, age and BMI. Participants with haplogroup H served as the reference group

Results: Participants included in this study had a mean age of 62,43 (sd 9,2), with a percentage of females of 54,1%. Haplogroups for the 1030 participants included H (n=419, 40,7%), J (n=88, 8,5%), T (109, 10,6%), Uk (n=236, 22,9%) and the less common haplogroups “others” (n=178, 17,3%).

Knees of participants with haplogroup J had significantly lower risk of large BMLs in lateral femur (OR=0,545;95%CI:0,299–0,996;p=0,048); in contrast, participants with haplogroup T had significantly increased risk of large BMLs in global tibia (OR=1,500;95%CI:1,020–2,206;p=0,039) and lateral plateau (OR=1,525;95%CI:1,003–2,320;p=0,048). No evidence of a protective or risk effect of any other haplogroup in the odds of large BMLs was detected in the rest of the knee compartments.

In terms of BMLs progression, participants with haplogroup J showed a lower risk of BML progression over time and an increased rate of no change in global femur compartment (OR=0,549;95%CI:0,309–0,977;p=0,041); participants with haplogroup T showed an increased rate of BML regression over time in global knee (OR=2,534;95%CI:1,236–5,195;p=0,011) as well as in medial plateau (OR=1,987;95%CI: 1,006–3,925;p=0,048). No other haplogroup was associated with the risk of BML progression for either knee region

Conclusion: mtDNA haplogroup J appeared to be OA protective, as it is associated with lower risk of large BMLs in lateral femur as well as a lower rate of BML progression over time in global femur. Interestingly, haplogroup T, despite its association with an increased risk of large BMLs, was significantly associated with BML regression rates in the whole knee. Both haplogroups J and T slow down the progression of BMLs

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/association-between-mitochondrial-dna-mtdna-haplogroups-and-bone-marrow-lesions-at-24-months-in-osteoarthritis-subjects-subjets-from-the-osteoarthritis-initiative/