Deoxycytidine + Deoxythymidine

An investigator-initiated Phase 2 trial at McGill University Health Centre (NCT04802707) is testing oral pyrimidine deoxynucleosides (deoxycytidine + deoxythymidine, dC/dT) specifically in patients with POLG-related mitochondrial DNA depletion syndromes, a rare disease group where mutations in the mitochondrial polymerase (POLG) cause children's mitochondria to lose DNA and their livers or brains to fail. [1][6] The same two nucleosides are the active ingredients in Kygevvi (doxecitine + doxribtimine), which the FDA approved on November 3, 2025 for a different mitochondrial disease: thymidine kinase 2 (TK2) deficiency. [2][3] The McGill trial asks whether the substrate enhancement approach helps POLG-related disorders, where the underlying defect (the DNA copying enzyme itself) is mechanistically distinct from TK2d. Six-month interim data on the first ten patients has already been published. [6]

Phase 2, single-center, open-label single-arm study sponsored by McGill University as an investigator-initiated trial enrolling patients with POLG-related disorders. [1][6] The compound itself is not novel: doxecitine (deoxycytidine) and doxribtimine (deoxythymidine) are the same molecules UCB already commercializes as Kygevvi, following FDA approval on November 3, 2025 based on an open-label Phase 2 program (NCT03845712, n=47) in TK2 deficiency. [2][3] The McGill trial has no FDA breakthrough, fast-track, or orphan designations independently attached to it in public registries; the parent compound carries priority review, breakthrough therapy, and rare pediatric disease designations for the TK2d indication. [3] Because this is an academic study in a distinct patient population, there is no formal regulatory submission timeline attached. Readout comes as investigator publication rather than a corporate press release: 6-month interim data on the first ten patients was published in 2024, [6] and data collection ran through at least December 2023, so a longer-term publication is plausible in the 2026 to 2027 window. Enrollment target is not disclosed in the registry summary. Any expansion of the approved Kygevvi label to POLG-related disorders would require UCB to run its own registrational study; academic signal alone does not extend a label (label expansion means FDA approval covering a broader set of patients than the originally approved use).

Mitochondria carry their own small DNA genome and constantly copy it to keep the cellular power plants running. To build that DNA they need deoxynucleoside building blocks, and TK2 is the enzyme that activates two of those building blocks (thymidine and deoxycytidine) inside the mitochondrion. When TK2 is broken by inherited mutations, mitochondria run short of raw material, lose DNA content, and muscle tissue collapses, usually in infancy or early childhood. Feeding patients large oral doses of dC/dT floods the pathway so alternate salvage enzymes can compensate: specifically, cytosolic dCK (deoxycytidine kinase) and TK1 (thymidine kinase 1) phosphorylate the excess substrate, and CMPK (cytidylate kinase) plus mitochondrial import of the resulting nucleotides can partially restore the mitochondrial dNTP pool. The genetic and preclinical case for TK2 deficiency is well established: knock-in mouse models and compassionate-use patient cohorts showed motor and survival gains before UCB commercialized the therapy. [3] For POLG-related disorders, which is the specific McGill trial population, the mechanistic case is subtly different. POLG mutations damage the mitochondrial polymerase itself, so extra substrate cannot fix a broken copying machine directly. The rationale for testing dC/dT anyway is that expanded dNTP pools may partially compensate for reduced polymerase fidelity or processivity; preclinical work suggested modest benefits, but this is a weaker mechanistic argument than the TK2d case. For OTHER MDDS subtypes not enrolled in this trial: DGUOK deficiency affects purine nucleosides (not pyrimidines), and MNGIE (TYMP) actually involves toxic accumulation of thymidine, where adding more dT would be actively harmful. Any benefit outside TK2d has to come from a broader substrate-pool effect that has not been rigorously demonstrated.

NCT04802707 is a single-center, open-label, single-arm Phase 2 study at McGill University Health Centre (Glen Site), enrolling patients aged 3 months to 60 years with biallelic pathogenic or likely pathogenic POLG variants. [1][6] The initial upper age limit was 18, later amended to 60 to expand eligibility. Data collection ran from October 14, 2021 through at least December 13, 2023, and 6-month interim results on the first ten patients (6 with Alpers-Huttenlocher syndrome, 2 with ataxia-neuropathy spectrum, 2 atypical POLG presentations) were published in 2024. [6] Specific primary endpoint definition and total enrollment target are not disclosed in the high-visibility registry summaries; the protocol tracks safety, tolerability, and clinical/biomarker efficacy consistent with prior open-label dC/dT programs. The design pattern for this class is well established from the UCB TK2d program: single-arm, with biomarkers of mitochondrial function (mtDNA copy number in tissue, meaning the count of mitochondrial DNA copies per cell as a proxy for mitochondrial health; respiratory chain enzyme activity), motor and hepatic function scales, and survival tracked over 12 to 24 months. [2] The obvious weakness for POLG is intra-genotype heterogeneity: Alpers-Huttenlocher, ataxia-neuropathy, and atypical POLG presentations have very different natural histories, and pooling them into a single arm makes attribution of any signal to genotype difficult. Diarrhea is the known dose-limiting toxicity for this drug class, driven by unabsorbed nucleoside reaching the colon; a parallel Phase 1 program (NCT06817590) at Boston Children's Hospital uses the same dC/dT combination in telomere biology disorders (where dT metabolism has been linked to telomere maintenance), and tolerability is the explicit primary endpoint there, making it the nearest ongoing safety comparator for this exact molecule class. [4] Renal impairment meaningfully alters PK (how the body absorbs, distributes, and clears the drug), which matters for the pediatric POLG population where kidney involvement can occur. [5]

Our model estimates a 83% chance this drug is eventually approved. It starts from the historical base rate for filing drugs in this area (about 93%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by a non-randomized design and its light or open-label blinding; it is held back by the sponsor's thin or weak approval record and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is the dominant concern. For POLG-related disorders, the biochemistry does not cleanly predict benefit: POLG mutations damage the polymerase itself, and no amount of substrate will fix a broken copying machine directly. The rationale rests on the hope that expanded dNTP pools partially rescue polymerase function, which is unproven at scale in humans. If enrolled patients span the Alpers-Huttenlocher, ataxia-neuropathy, and atypical POLG subtypes, the trial risks a null result driven by intra-genotype heterogeneity rather than clear drug failure. Safety risk is more manageable: dose-limiting diarrhea is the known signal for the drug class, [4] and renal impairment shifts PK enough to matter in the pediatric population. [5] Execution risk: investigator-initiated academic trials in ultra-rare disease routinely stall on enrollment, and there is no commercial sponsor pressure to accelerate. Commercial risk is the sharpest. Even if the trial shows a signal, UCB owns the compound and the manufacturing, and academic data cannot expand the Kygevvi label without a company-sponsored registrational study. Off-label use in non-TK2d MDDS would depend on payer willingness, and Kygevvi's US wholesale acquisition cost is $16,438.50 per carton with weight-based dosing, [7] a pricing structure that has drawn scrutiny even for the approved TK2d indication. Whether McGill or the trial investigators hold any IP or licensing agreement with UCB that could create a commercialization path for POLG data is not publicly disclosed; absent such an arrangement, positive academic data is unlikely to translate into a labeled option for POLG patients on any near-term timeline.

Noise for commercial investors, real for the mitochondrial disease community. The commercial story on this mechanism was already written when UCB obtained approval for Kygevvi in TK2 deficiency (November 3, 2025); [3] the McGill trial cannot generate a stock-moving event because there is no listed sponsor with equity exposure to the outcome. The 6-month interim publication [6] shows the class is tolerated in POLG patients, which is a real scientific data point but not commercially actionable on its own. Watch this only if a genotype-stratified subgroup analysis (results separated by the specific POLG mutation each patient carries) shows a clean efficacy signal in Alpers-Huttenlocher or ataxia-neuropathy patients, because that would give UCB a specific expansion target worth their own registrational investment. The single data point that would elevate this from academic curiosity to signal: a documented mtDNA copy-number recovery in tissue biopsy paired with motor or hepatic function gains over 12+ months in a POLG cohort. Given data collection ran through late 2023, a longer-term readout publication is plausible in 2026 to 2027. Check back after any publication from the McGill group on longer-term outcomes, and cross-reference against UCB investor communications for any hint of a company-sponsored POLG expansion study. Absent that, this is a small academic effort answering a scientifically valid but commercially inert question.