Valganciclovir

Limits on what this shows

• Null findings: [13] above. No adult tissue or biopsy data found.
• At-risk subgroups: None identified. CMV-infected cells are the label's site of phosphorylation [1].

Regulatory status

No mitochondrial or lactic-acidosis statement in the label text read; the boxed warning covers hematologic toxicity, fertility, fetal toxicity and carcinogenicity [1]. The label states ganciclovir is phosphorylated in CMV-infected cells [1].

Population evidence

Human, valganciclovir in the cohort: Ortiz-Gracia 2022 [13], 34 infants with congenital CMV (16 symptomatic infants on valganciclovir 32 mg/kg/day divided twice daily for 6 months, 18 untreated) and 8 healthy controls, peripheral blood mononuclear cells at baseline, 3 to 6 and 12 months: complex IV and citrate synthase activities and mtDNA levels did not differ from controls (p above 0.05); mtDNA/nDNA about 80 to 98 against a control reference range of 60 to 144. The authors cite the small sample as a limit. Models (ganciclovir, not valganciclovir; cells that express viral thymidine kinase, HSV-TK): adenoviral HSV-TK plus ganciclovir in rats caused hepatocyte death, enlarged mitochondria with disrupted cristae and a marked fall in liver mtDNA, and ganciclovir was incorporated into nuclear and mtDNA in transduced rat hepatoma cells [14]; ganciclovir nucleotides were 10 to 30% of the mitochondrial nucleotide pool by 48 h, with lower mtDNA, reduced membrane potential and liver dysfunction [15]; in HSV-TK human cells higher ganciclovir concentrations killed non-dividing senescent cells via mtDNA damage [16]; TK/ganciclovir disrupted membrane potential and released cytochrome c in transfected cells (PMID 10866313, listing summary) [17].

Mechanism

Named but conditional: initial phosphorylation by viral thymidine kinase, then incorporation into mtDNA and loss of mtDNA [14][15]. Whether this occurs in uninfected human tissue during valganciclovir treatment is not established; the human infant study is the only test and it is null [13].

Counter-evidence held on record

Framing caution for the consumer surface Do not present this as mitochondrial damage in patients: the only human data are null and the positive data use gene-therapy models. No supplement is implied.

Limits on what this shows

• Null findings: In a CMV-infected fibroblast study ganciclovir left CMV-induced inflammatory markers unchanged and oxidative stress was supplied by H2O2, not measured as a ganciclovir effect (PMID 10702726, listing summary) [26]. The two animal results above point in opposite directions, so neither can be read as a drug effect on antioxidant status.
• At-risk subgroups: None identified.

Regulatory status

No oxidative-stress or antioxidant statement in the label text read [1].

Population evidence

(i) Tier B, ganciclovir (not valganciclovir), mouse: Lv 2014 [32], BALB/c mice injected intravenously with HCMV, groups control, model, model plus ganciclovir (positive control) and model plus curcumin at three doses. Ganciclovir and curcumin both reduced HCMV IgM and viral DNA load, lowered AST, ALT, CK, LDH, TNF-alpha and IL-6, reduced malondialdehyde and raised superoxide dismutase and glutathione relative to the infected model group. The record read (PubMed paraphrase) gave no doses, animal numbers or effect sizes. Because ganciclovir lowered viral load, this is less infection-driven oxidative stress, not evidence of a direct antioxidant action. (ii) Tier C CLASS EXTRAPOLATION, acyclovir (not ganciclovir or valganciclovir; neither was in either cohort): 40 male Wistar rats, acyclovir 150 mg/kg intraperitoneally daily for 7 days in an acyclovir nephrotoxicity model [34]; kidney MDA 0.25 to 1.46 nmol/mg protein, GSH 14.32 to 3.05 umol/mg, catalase 29.92 to 11.42 U/mg, GPx 17.03 to 4.12 U/mg, SOD 29.87 to 10.40 U/mg (n=5 per group, p below 0.001 versus control; reversed by N-acetylcysteine). Separate rat study [35], 18 male Wistar rats, acyclovir 10 and 40 mg/kg/day orally for 21 days: at 40 mg/kg serum MDA 0.44 to 1.44 uM/g and serum SOD 4.58 to 1.82 U/mg. Both rat studies used a nephrotoxic or testicular-toxicity framing, and no human-equivalent exposure comparison was made here. (iii) Human: a PubMed search of (valganciclovir or ganciclovir) with glutathione, superoxide, malondialdehyde or antioxidant listed 14 of 60 records, and none measured oxidative markers in patients treated with either drug [37]. The mitochondrial HSV-TK membrane-potential and cytochrome c work [15][17] did not measure reactive oxygen species.

Mechanism

Absent as a demonstrated pathway for valganciclovir or ganciclovir. Mitochondrial dysfunction in thymidine-kinase-expressing cells [15][17] is a possible route to oxidative stress but was not measured as such. For acyclovir the rat authors frame the changes as nephrotoxic oxidative injury [34]; whether ganciclovir behaves the same way was not tested.

Counter-evidence held on record

Framing caution for the consumer surface Do not present this as antioxidant depletion or as an antioxidant benefit of valganciclovir. The direction is unresolved, the only ganciclovir data are in mice, and no antioxidant supplement is implied.

Limits on what this shows

• Why no grade: Dual condition met. Channels tried: PubMed "valganciclovir copper" (client error, then not repeated verbatim), "ganciclovir copper" (3 records), "valganciclovir zinc" (copper not reported in the one clinical hit) [28]; in the repair pass the listing (valganciclovir OR ganciclovir) AND (copper OR ceruloplasmin) was run in full and returned the same 3 records, none measuring serum or plasma copper or ceruloplasmin: aqueous humor trace elements in CMV retinitis, where copper was higher in disease eyes (PMID 31065104, a local disease effect, not a systemic copper measure, and ganciclovir exposure not stated in the abstract), an artemisinin-hybrid chemistry paper (PMID 32485071), and a DNA-synthesis paper using copper as a reagent (PMID 7619825) [37]. Ceruloplasmin appeared in no record. WebSearch for ganciclovir serum copper and ceruloplasmin returned only unrelated ceruloplasmin references. A class-level search (acyclovir or other antiviral nucleoside analogues with copper) was not run because the WebSearch budget was exhausted. No label statement [1]. No circulating claim found [29][30]. Classification: genuine gap in the literature found for this drug, with the class-level channel untried.

Limits on what this shows

• Why no grade: Dual condition met. Channels tried: PubMed "valganciclovir ferritin" (6 records) [28]; in the repair pass the listing valganciclovir AND (ferritin OR iron deficiency OR hepcidin) returned 7 records [37]: a kidney-transplant iron-loss study (PMID 19076330, 39 recipients, ferritin below 200 ng/mL in 51%, baseline hemoglobin 11.8 g/dL; valganciclovir appears only as an indexing term and not in the abstract), a dasatinib-treated leukemia case series in which CMV hemorrhagic colitis presented as iron deficiency anemia and was managed with valganciclovir plus iron (PMID 39494242, blood loss from CMV colitis), CMV-associated hyperferritinemia and hemophagocytic picture case reports (PMIDs 35847599, 41064724, 41450345; ferritin 1,570 ng/mL in one), and a roxadustat report in kidney recipients with CMV-associated anemia despite valganciclovir (PMID 41644404; hemoglobin 8.9 g/dL, ferritin 673 ng/mL). In each the iron signal belongs to CMV disease or colitis, not to the drug. The lung-transplant pancytopenia case [31] recorded ferritin 1,976 ng/mL (high, with transfusion and inflammation possible) and the authors suggest folate deficiency impaired iron use; no iron deficit and no attribution to valganciclovir. A second listing (valganciclovir or ganciclovir with hepcidin, transferrin, serum iron or iron deficiency) returned HTTP 429 twice and was not read; WebSearch budget was exhausted before a second phrasing could be run. No circulating claim found in pages opened [29][30]. Classification: tooling-limited, not a confirmed gap; a cohort with iron indices before and during valganciclovir may exist in records that could not be listed.

Limits on what this shows

• Null findings and competing explanations: Transplant recipients carry dysbiosis from immunosuppressants, antibiotics, surgery and kidney disease, which would swamp any valganciclovir signal; the BK study did not isolate ganciclovir [36]. Infection itself and IVIG shifted the mouse microbiome independently of acyclovir [33]. A third paper on dysbiosis in renal transplant recipients (PMC7074359) was not opened; the PLOS ONE version of [33] was not read (HTTP 429).
• At-risk subgroups: None identified for valganciclovir.

Regulatory status

No microbiome statement in the label text read; diarrhea occurred in 30 to 41% of patients in label trials, equal to the oral ganciclovir comparator in the transplant trial (30% vs 29%) [1].

Population evidence

Cohort test: valganciclovir and ganciclovir were not in the cohort of the class source. Ramakrishna 2020 [33], read as the bioRxiv preprint: C57BL/6 mice, HSV-1 infected by corneal scarification or uninfected, each given PBS, acyclovir (50 mg/kg intraperitoneally daily for 3 days in Methods; the Results text gives 1.25 mg/mouse, an unresolved inconsistency), IVIG, or both; n=8 per Methods (per-group n only in figure legends not seen); one fecal collection on day 7; 16S rRNA V4 to V5 sequencing. All four treatment groups differed in community structure (Adonis P below 0.05); acyclovir produced "more drastic shifts", reducing Bacteroidetes taxa in infected males and Blautia taxa in infected females (FDR-adjusted P below 0.05) and lowering Akkermansia muciniphila; in uninfected females acyclovir raised Turicibacter. Sex-specific, single time point, and the acyclovir effects on the main taxa were reported in infected animals. Human context, valganciclovir not analyzed: a 16S case-control study of 25 BK-virus-infected and 23 control kidney transplant recipients gave ganciclovir prophylaxis to most (Methods states all; Table 1 lists 15 of 25 vs 18 of 23, p=0.173) together with low-dose sulfamethoxazole/trimethoprim, and the microbiome comparison was by BK infection, with ganciclovir not analyzed as a determinant [36]. Earlier searches found only case reports with fecal transplant and mouse tool-use studies [25].

Mechanism

Absent for valganciclovir. No fecal or luminal drug concentration or in vitro bacterial susceptibility data were retrieved. Diarrhea is a clinical event, not a microbiome measurement.

Counter-evidence held on record

Framing caution for the consumer surface Class extrapolation from a mouse study of a different drug. It must not be shown as evidence that valganciclovir disrupts gut bacteria, and no probiotic or other supplement is implied.

Limits on what this shows

• Null findings and competing explanations: No non-use comparator for hemoglobin was found, and anemia is common after transplant for other reasons (surgery, kidney disease, mycophenolate, which the label lists as needing hematologic monitoring), so label anemia rates are not all attributable to valganciclovir. Extending prophylaxis from 100 to 200 days (N=326, D+/R- kidney) produced similar adverse event rates in the abstract [5]. A post-transplant case with severe neutropenia (ANC 48/uL) did not recover after valganciclovir, voriconazole and acitretin were stopped and normalized only after vitamin B6 was replaced [18], an example of a competing, non-drug cause. Valganciclovir overdosing relative to renal function was not significantly associated with leukopenia or neutropenia in one cohort [10].
• At-risk subgroups: Renal impairment (half-life up to 67.5 h at CrCl 10 mL/min or below) [1]; infants and children (higher neutropenia and anemia incidence) [1]; baseline ANC below 1000 or pre-existing cytopenias [1]; co-use of mycophenolate and other myelosuppressive drugs [1]; probenecid (ganciclovir AUC up 53%, wide range) [1]; longer courses and higher dose [6][11].

Regulatory status

Boxed warning [1]: severe leukopenia, neutropenia, anemia, thrombocytopenia, pancytopenia and bone marrow failure including aplastic anemia. Section 5.1: avoid if ANC below 500 cells/uL, platelets below 25,000/uL or hemoglobin below 8 g/dL; caution with pre-existing cytopenias, myelosuppressive drugs or radiation; "cell counts usually begin to recover within 3 to 7 days after discontinuing drug"; frequent CBC with differential and platelets, especially in infants, renal impairment and baseline neutrophils below 1000; growth factors may be considered. Label tables [1][3]: CMV retinitis maintenance (N=370, Table 4) hemoglobin below 6.5 g/dL 7%, 6.5 to below 8 13%, 8 to below 9.5 16%; ANC below 500 19%. Adult solid organ transplant (Table 7; valganciclovir N=244 vs oral ganciclovir N=126): hemoglobin below 6.5 1% vs 2%, 6.5 to below 8 5% vs 7%, 8 to below 9.5 31% vs 25%; ANC below 500 5% vs 3%. Kidney transplant (Table 8): hemoglobin bins 0%, 5%, 17% at Day 100 (N=164) and 1%, 1%, 15% at Day 200 (N=156); ANC below 500 9% and 10%. Pediatric: anemia among the most frequent reactions (over 10%) through Day 100 and more frequent than in adults; severe neutropenia 30% (17/57) with treatment to Day 200 vs 5% (3/63) to Day 100 in pediatric kidney transplant. Tables were read through summarizing fetch tools and the two summaries of the CMV retinitis table differed slightly in their anemia wording; bin figures above are from the MedLibrary copy [3].

Population evidence

(a) Label trials above: hemoglobin directly measured with this drug in two independent populations (CMV retinitis, solid organ transplant). The comparator for anemia is active (oral ganciclovir) or a shorter course, not non-use. (b) Boston cohort, N=1,650 adult solid organ transplant recipients (343 lung, 200 heart, 1,107 kidney), valganciclovir median exposure 182 days (IQR 108 to 258) [8]: leukopenia, severe leukopenia (WBC below 1,500) and thrombocytopenia were more frequent with valganciclovir than without (log-rank p below 0.001 each) and G-CSF use was higher; anemia was defined (Hgb below 12 g/dL women, below 13.5 men, or fall over 1 g/dL) but its incidence was not in the abstract-level text read. (c) TriNetX kidney cohort, N=15,398, valganciclovir and/or ganciclovir within 30 days of transplant [9]: neutropenia and/or leukopenia (ANC below 1000 and/or WBC below 3500) in 74.84% in year 1; anemia was not reported. (d) Bixby 2021, liver transplant, N=96 [11]: ANC below 1000 in 60% (21/35) on 900 mg/day for 180 days vs 10% (6/61) on 450 mg/day for 90 days, OR 12.2 (95% CI 4.17 to 35.9); historical comparison, so duration and era are confounded. (e) Meta-analysis, 7 observational kidney cohorts, N=1,431 [12]: leukopenia 450 vs 900 mg/day OR 0.65 (0.34 to 1.22), I2 74%. (f) Hammer 2024, N=131 kidney recipients analyzed [10]: dosing above renal-function-appropriate level vs not, leukopenia OR 2.28 (0.49 to 10.48) and neutropenia OR 2.45 (0.28 to 21.62), both non-significant; leukopenia rose with follow-up weeks (OR 1.54, 1.31 to 1.80). (g) Congenital CMV, randomized, N=96 randomized, 86 evaluable [7]: grade 4 neutropenia 19% in the first 6 weeks, then 21% (6 months) vs 27% (6 weeks), P=0.64. (h) Cochrane review of 37 trials, 4,342 participants [6]: leukopenia more common with ganciclovir than aciclovir, more common with longer prophylaxis; valganciclovir and ganciclovir did not differ in efficacy or adverse effects across three small studies. Sources (b) to (h) were read as abstracts, reconstructed OpenAlex abstracts or summarizing-fetch PDF extractions; no hemoglobin effect size with a CI and a non-exposed comparator was retrieved.

Mechanism

Toxicity of the drug to marrow, tied by the label to the drug (recovery in 3 to 7 days after stopping). The usual explanation, inhibition of DNA synthesis in rapidly dividing marrow cells by the ganciclovir nucleotide, is the standard pharmacology account, but no primary human progenitor study was retrieved in this review, so the mechanism is stated as presumed, not as sourced.

Why orange, not red or yellow

Category-endpoint convention, as used for dapsone, primaquine and ribavirin: the Step 3 category "oxygen delivery/utilization" is graded here as a drug-toxicity endpoint, with direction labeled toxicity. Orange is justified because (i) it is a boxed-warning reaction on the regulatory label, (ii) hemoglobin was directly measured in two independent label trial populations with threshold incidences, (iii) the boundary is definable (window, renal function, age, co-drugs). It is held below red because it is a drug toxicity (Hard Rule 2), because no non-use hemoglobin comparator was retrieved, and because the mechanism is presumed rather than sourced.

Counter-evidence held on record

Framing caution for the consumer surface This is a drug side effect monitored by blood counts. Anemia or low neutrophils on valganciclovir should be reported to the prescriber, who adjusts dose or interrupts treatment. Taking iron, B12 or folate on one's own is not supported as a remedy by any source found here, and iron supplements could mask a different cause.

Limits on what this shows

• Null findings: [13]: respiratory chain and TCA-cycle activity proxies unchanged in infants.
• At-risk subgroups: None identified.

Regulatory status

No energy-metabolism statement in the label text read [1].

Population evidence

Same sources as Mitochondrial burden (see Overlap map): membrane potential fell with ganciclovir nucleotides in mitochondria [15]; TK/ganciclovir disrupted membrane potential in transfected cells [17]; infant complex IV/citrate synthase preserved [13]. Searches for ATP, respiration, oxygen consumption and OXPHOS with ganciclovir or valganciclovir found no direct measurement.

Mechanism

Indirect: loss of mtDNA would reduce respiratory chain subunits [14][15]; not demonstrated for ATP in any model found.

Counter-evidence held on record

Framing caution for the consumer surface Not evidence of low cellular energy in people taking valganciclovir. Fatigue on valganciclovir may reflect anemia, infection or the underlying illness.