What genetic tests (like DIO2) could predict benefit from adding T3?

The short version: the DIO2 Thr92Ala polymorphism (rs225014) is the best-studied candidate, but the evidence that it predicts who benefits from adding T3 is thin and contradictory.

The DIO2 gene makes the enzyme that converts T4 into T3 inside cells, especially in the brain. A common variant, the Thr92Ala polymorphism (rs225014), is carried by about 15% of the population [2]. The idea is that people with this variant convert T4 less efficiently and would do better with direct T3. A 2012 review [2] cites the WATTS study, which found that patients with this polymorphism had better outcomes when their T4 dose was reduced and T3 was added. That is the main positive signal.

But the evidence since then has not been clean. A 2025 study in Chinese thyroid cancer patients [1] looked at the same DIO2 variants (rs225014 and rs12885300) and found that carriers of the mutant genotype needed higher T4 doses to suppress TSH, but after six months there was no significant effect on TSH, FT3, or FT4 levels. The genotype did not predict who would benefit from a different T4/T3 balance in that study. A 2021 study in Brazilian children with autism [18] found the Thr92Ala variant was not more common in autism and did not predict treatment response, though carriers had slightly better adaptive behavior scores.

The retrieval also includes other genes that might matter. A 2008 study [15] found that variants in the OATP1C1 transporter, which moves thyroid hormone into the brain, were linked to fatigue and depression in hypothyroid patients on T4 alone. That is a different mechanism—not conversion, but transport—and it has not been tested in a trial of adding T3. A 2003 study [14] found that DIO1 and DIO2 SNPs were associated with differences in TSH and rT3 levels in healthy people, but those are small effects and have not been linked to T3 treatment response.

The problem is that no trial has randomized patients by genotype and measured hard outcomes. The WATTS study is the closest, and it is a subgroup analysis from a single trial. The 2025 study [1] contradicts the idea that DIO2 genotype predicts T4 dose requirements in a straightforward way. The evidence base is too small and too mixed to recommend genetic testing as a reliable guide.

My call: genetic testing for DIO2 or OATP1C1 variants cannot currently predict who will benefit from adding T3. The mechanism is plausible, the evidence is not. Confidence: low.