Oncofusion

start / graph / drug-vorinostat

Drugstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q39-acetylation-cycle

Vorinostat (SAHA) - the only HDAC inhibitor given to newly diagnosed infants with ALL, in a 50-patient trial whose endpoint was tolerability and which reported four treatment-related deaths, all infectious

also called SAHA · suberoylanilide hydroxamic acid · Zolinza · MK-0683

Vorinostat is an approved drug that blocks the enzymes which strip acetyl tags off histones. It has been given to fifty newly diagnosed infants with acute lymphoblastic leukaemia, alongside bortezomib and standard chemotherapy - the only HDAC blocker ever taken that far in this exact population. The trial was designed to ask whether infants could tolerate it, not whether it worked, and four children died of infection during treatment.

The same thing, for a scientist

Vorinostat is a hydroxamic-acid pan-HDAC inhibitor, FDA-approved for cutaneous T-cell lymphoma, that inhibits class I and class II enzymes including HDAC1, 2, 3 and 6 [E3]. **NCT02553460 ('TINI', Total Therapy for Infants with ALL I, St. Jude) is the reason this entry exists**: a single-arm phase 1/2 study at 18 hospitals in the USA and Canada that enrolled **50 newly diagnosed infants aged 1 year or younger, 35 of them KMT2A-rearranged**, incorporating bortezomib and vorinostat into a dexamethasone / mitoxantrone / pegaspargase induction and reinduction backbone [E1][E2]. The preclinical rationale named by the investigators is that bortezomib and vorinostat were previously identified as **active agents against KMT2A-rearranged leukaemia** [E1]. **The primary endpoint was completing induction and reinduction with no protocol-defined dose-limiting toxicities - tolerability, not efficacy** [E1]. Vorinostat was escalated through 100, 150 and 180 mg/m2; no DLTs occurred at the first two levels and 44 patients received dose level 3, at which four induction DLTs and one reinduction DLT occurred [E1]. Median follow-up for surviving patients was 4.9 years. **Four patients had treatment-related deaths, all infectious** (parainfluenza, cytomegalovirus x2, Aspergillus fumigatus), three of them **after** induction and reinduction, in consolidation and maintenance [E1]. The investigators' own interpretation is that adverse events were consistent with previous infant ALL studies and that *"infections remain a substantial cause of morbidity and mortality in this immunocompromised patient population and extended beyond the induction and reinduction chemotherapy phases"* [E1]. **This entry makes no efficacy claim, because the trial did not test one**: there was no randomised comparator, the KMT2A-rearranged and germline patients were enrolled together, and vorinostat was given with bortezomib inside a chemotherapy backbone, so nothing in the design can attribute any outcome to HDAC inhibition.

Go deeper
What this trial can and cannot support

Can support:

  • HDAC inhibition has been delivered to newly diagnosed infants with ALL at 180 mg/m2/day on a 4-days-on schedule inside intensive chemotherapy, and the toxicity profile did not exceed what infant ALL regimens already produce 1.
  • The trial is registered, completed accrual (2016-01-29 to 2021-11-17), has posted results, and this Lancet Haematology report is the final report 12.
  • The agent reached the population this graph exists to serve. Almost nothing else in the acetylation layer has.

Cannot support:

  • Any efficacy claim whatever. Single-arm, primary endpoint tolerability, no HDAC-inhibitor-free comparator arm.
  • Any attribution to HDAC inhibition. Bortezomib, a proteasome inhibitor, was given in the same blocks. A two-agent addition to a backbone cannot be decomposed.
  • Any attribution to an HDAC enzyme. Vorinostat is pan-class. This is the central problem of the whole beat; see ana-no-hdac-inhibitor-result-in-this-disease-attributes-to-an-enzyme.
  • Any claim restricted to KMT2A-rearranged disease. 35 of 50 were KMT2A-rearranged; the study enrolled infants with and without the rearrangement by design 1.

Note that the registered enrolment is 50 and the trial's overall status is ACTIVE_NOT_RECRUITING with an estimated completion of October 2031, because long-term follow-up continues 2. The treatment phase is finished; the record is not closed.

An HDAC inhibitor in this disease has at least two mechanisms, and an entry naming only histone acetylation is incomplete

Three distinct routes by which an HDAC inhibitor acts in KMT2A-rearranged leukaemia are already recorded in or adjacent to this graph, and they are not the same event:

  1. Bulk histone acetylation. The assumed mechanism, and the one the drug class is named for.
  2. Transcriptional suppression of the RNF20/RNF40/WAC E3 ligase complex, depleting H2B ubiquitination. This is the route mechanism-panobinostat-depletes-h2bub-mll-r-all records for panobinostat - indirect, downstream of HDAC inhibition, and reaching an axis no other clinical-stage molecule reaches.
  3. Acute release of free P-TEFb from the 7SK snRNP. mechanism-ptefb-extraction-from-the-7sk-reservoir and gene-hexim1 establish that SAHA - vorinostat itself - releases P-TEFb acutely, with HEXIM1 induction following as the feedback that rebuilds the particle 4. In a disease whose core lesion drives transcriptional elongation, an agent whose first action is a transient surge of free elongation factor is doing something the histone-acetylation story does not predict, and possibly something in the wrong direction for the first hours of dosing.

Route 3 is specifically a vorinostat finding, not a class generalisation borrowed from another compound, which is why it belongs in this entry rather than only in the P-TEFb beat. An HDAC inhibitor entry that names only histone acetylation is incomplete, and the graph should not let the drug's name stand in for its mechanism.

Bounded rejection: what was searched and not used

(MLL[tiab] OR KMT2A[tiab]) AND (vorinostat[tiab] OR SAHA[tiab] OR "suberoylanilide"[tiab]) returns 8 records (2026-09-20). Title-checked, the usable ones are PMID 41692013 (this trial) and 26447190 (the genetic HDAC dissection, cited in gene-hdac3). PMID 35571127 concerns a novel compound 'I1' rather than vorinostat; 32917124, 29567781, 24522270 and 20701602 were judged off-beat on title and not read further; 40437770 is a revumenib review.

Deliberately not written as separate entries in this beat: belinostat and givinostat, for which no KMT2A-rearranged or infant result was found; and the HDAC inhibitor 'AN-9' (PMID 12384433, Blood 2002), which reports selective toxicity to acute leukaemia and drug-resistant lines but was not pursued clinically and has no KMT2A-specific data. A future cycle re-running this beat should know these were seen and dropped, not missed.

Facts and figures
targetgene-hdac3
target notePan-HDAC. Vorinostat inhibits class I and class II enzymes including HDAC1, HDAC2, HDAC3 and HDAC6. gene-hdac3 is named in the schema-required target field because it is the best genetically attributed HDAC in this disease, NOT because vorinostat is selective for it. It is not.
modalitysmall molecule, hydroxamic acid, orally available, zinc-chelating pan-HDAC inhibitor
statusFDA approved for cutaneous T-cell lymphoma; investigational in infant ALL
chembl noteChEMBL was not queried for this entry; a ChEMBL activity-row count would in any case measure medicinal-chemistry effort rather than tractability, per this project's database-scope rule.
efficacy claimNONE - the trial did not test efficacy and this entry makes no efficacy claim
attribution to an hdac enzymefalse
trials
[
  "NCT02553460"
]
trial detail
{
  "nct": "NCT02553460",
  "acronym": "TINI",
  "phase": "1/2",
  "design": "single-arm, interventional, 18 sites USA and Canada",
  "sponsor": "St. Jude Children's Research Hospital",
  "collaborators": [
    "Gateway for Cancer Research",
    "Baylor College of Medicine"
  ],
  "n": 50,
  "n_kmt2a_rearranged": 35,
  "population": "newly diagnosed infants aged 1 year or younger with ALL or undifferentiated leukaemia, with and without KMT2A rearrangement",
  "median_age_days": 171,
  "primary_endpoint": "completion of induction and reinduction with no protocol-defined dose-limiting toxicity (TOLERABILITY, not efficacy)",
  "vorinostat_doses_mg_m2": [
    100,
    150,
    180
  ],
  "n_at_dose_level_3": 44,
  "dlts_at_dl3": "4 induction, 1 reinduction",
  "median_followup_years": 4.9,
  "status": "ACTIVE_NOT_RECRUITING (long-term follow-up; treatment accrual 2016-01-29 to 2021-11-17)",
  "final_report": "PMID 41692013"
}
key ae
[
  "grade 3-4 during induction: hypertension 48%, infection 44%, fever and neutropenia 42%, hypokalaemia 34%, hypocalcaemia 34%, anorexia 30%, ALT elevation 20%, hypoalbuminaemia 20%",
  "most frequent during reinduction: hypertriglyceridaemia 23% of 44",
  "four treatment-related deaths, all infectious: parainfluenza (induction), cytomegalovirus (consolidation and maintenance), Aspergillus fumigatus (maintenance)"
]
additional mechanisms
[
  "acute release of free P-TEFb from the 7SK snRNP, with HEXIM1 induction as feedback (a vorinostat/SAHA-specific finding)",
  "for the class: transcriptional suppression of RNF20/RNF40/WAC depleting H2B ubiquitination, recorded for panobinostat"
]
resistance
[
  "not characterised in this population; no resistance mechanism has been reported for HDAC inhibition in KMT2A-rearranged leukaemia",
  "any allele-directed component of the class I mechanism would be defeated by loss of the wild-type KMT2A allele"
]
How this could bite us back
  • Four of fifty infants died of treatment-related infection, and three of those deaths came after the HDAC inhibitor had stopped. In a population this small and this immunologically immature, an agent that adds any immunosuppression at all needs an efficacy argument that this trial was not designed to provide.
  • Vorinostat is pan-class. Every physiological process that depends on reversible lysine acetylation is affected, in a one-year-old whose haematopoiesis, gut epithelium, thymus and brain are all in active development. Isoform-selective agents were developed precisely to escape this, and the two isoform-selective classes tested in this disease have either failed in vivo (class IIA) or proved non-fusion-specific (HDAC3).
  • Because vorinostat was given with bortezomib inside a chemotherapy backbone, a positive outcome would be attributed to HDAC inhibition by readers and could not be by the data. That is a mechanism for a wrong belief to become established, and it is worth naming before any follow-on trial is designed.
  • The only efficacy-endpoint trial of any HDAC inhibitor in acute leukaemia in this graph was terminated for lack of efficacy. Tolerability in infants and futility in adults are compatible, and the field should not treat the former as encouragement about the latter.
  • If the acute action of vorinostat is to release free P-TEFb, then in a leukaemia whose lesion is excess transcriptional elongation the first hours of each dose may push in the wrong direction. Nobody has measured this, and a schedule optimised on histone acetylation would be blind to it.
Still open
  1. Did the KMT2A-rearranged and germline infants in TINI differ in outcome? The abstract does not say, and the full text and the posted registry results would answer it.
  2. Can the vorinostat contribution be separated from the bortezomib contribution in any analysis of this trial, or is the two-agent addition permanently entangled?
  3. Was any pharmacodynamic marker of HDAC inhibition - histone acetylation in blasts, HEXIM1 induction, H2B ubiquitination - measured in these 50 infants? If not, the trial cannot say whether the target was engaged.
  4. Three of the four treatment-related deaths were after induction and reinduction, in consolidation and maintenance, when vorinostat had stopped. Does an HDAC inhibitor given early leave a durable immunosuppressive effect, and is that measurable?
  5. Does vorinostat release P-TEFb in a KMT2A-rearranged cell, and if so does the first day of dosing transiently accelerate the fusion's own elongation programme?
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    NCT02553460 was a single-arm phase 1/2 study at 18 hospitals in the USA and Canada that enrolled 50 newly diagnosed infants aged 1 year or younger with acute lymphoblastic or undifferentiated leukaemia, 35 of whom had KMT2A rearrangements, incorporating bortezomib and vorinostat (dose-escalated through 100, 150 and 180 mg/m2 orally once daily on days 1-4, 8-11 and 15-18) into a dexamethasone, mitoxantrone and pegaspargase induction and reinduction backbone; the primary endpoint was completing induction and reinduction with no protocol-defined dose-limiting toxicities; no DLTs occurred at the first two vorinostat dose levels, 44 patients received dose level 3, at which four induction DLTs and one reinduction DLT occurred; median follow-up for alive patients was 4.9 years; and four patients had treatment-related deaths, all infectious (parainfluenza during induction, cytomegalovirus in consolidation and in maintenance, and Aspergillus fumigatus in maintenance).

    PMID 41692013DOI
    What this evidence cannot show

    Single-arm with no HDAC-inhibitor-free comparator, so nothing about efficacy can be concluded and this entry claims none. Vorinostat was given together with bortezomib, so the two agents cannot be separated. Infants with and without KMT2A rearrangement were enrolled together by design. The primary endpoint is tolerability. Only the structured abstract was read for this entry; PMCID PMC12998465 exists and the full text was not retrieved, so the efficacy outcomes that the paper reports beyond the abstract are not represented here and a later reader should fetch them rather than assume this entry is complete. The abstract reports adverse-event frequencies inclusive of all three dose levels, which understates rates at the dose actually used in expansion.

    Evidence that argues against this (1)
    • The only randomised-quality clinical fact about an HDAC inhibitor in acute leukaemia in this graph is negative: NCT00723203, a phase 2 of panobinostat in relapsed or refractory ALL and AML, enrolled 16 adults and was TERMINATED with the registry stating 'Terminated early due to a lack of efficacy'. A tolerability trial in infants does not overturn that, and the two facts should be read together rather than separately. City of Hope Medical Center; National Cancer Institute 2010
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry

    ClinicalTrials.gov record NCT02553460, organisation study id TINI, 'Total Therapy for Infants With Acute Lymphoblastic Leukemia (ALL) I', lead sponsor St. Jude Children's Research Hospital with collaborators Gateway for Cancer Research and Baylor College of Medicine, is an interventional single-group phase 1/2 treatment study with actual enrolment 50, actual start 2016-01-29, actual primary completion 2022-05-10, estimated completion October 2031, overall status ACTIVE_NOT_RECRUITING as of a 2026-06 status verification, with results first posted 2023-06-07; the intervention list includes vorinostat and bortezomib alongside ITMHA, dexamethasone, mitoxantrone, pegaspargase, asparaginase Erwinia chrysanthemi, cyclophosphamide, mercaptopurine, methotrexate, leucovorin calcium, cytarabine, etoposide and vincristine.

    Total Therapy for Infants With Acute Lymphoblastic Leukemia (ALL) I · St. Jude Children's Research Hospital · ClinicalTrials.gov API v2, retrieved 2026-09-20 via the raw API per this project's registry rule 2026
    NCT02553460
    What this evidence cannot show

    The raw API was used rather than the MCP tool, per this project's standing rule, and the whyStopped field is not applicable because the study was not stopped. The registry's enrolment figure and the publication's agree at 50. Registry intervention lists describe the whole protocol including maintenance and are not a statement of what any individual patient received. The estimated completion date of 2031 reflects long-term follow-up, not continuing treatment.

    Evidence that argues against this (1)
    • A registry record is a PLAN, not a result. Registration states what a sponsor intended to do; it does not establish that the trial completed, that it enrolled the KMT2A-rearranged subgroup in usable numbers, or that any effect was seen - and this beat's clearest cautionary case is an HDAC-directed agent that eradicated KMT2A-rearranged ALL in vitro and then failed in vivo at the maximum achievable dose. A registry entry cited as evidence of activity is citing the hypothesis, not the test. Verbeek TCAI et al. 2024
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Vorinostat, romidepsin, belinostat and panobinostat are FDA-approved for haematological malignancies and inhibit class II and/or class I HDACs, including HDAC1, 2, 3 and 6.

    Functional-genetic dissection of HDAC dependencies in mouse lymphoid and myeloid malignancies. · Matthews GM, Mehdipour P, Cluse LA, Falkenberg KJ, Wang E, Roth M, et al. · Blood 126(21):2392-2403 2015
    PMID 26447190DOI
    What this evidence cannot show

    This is a statement of the approved agents' isoform coverage taken from the introduction of a primary paper rather than from a pharmacology review or a label; it is cited for the multi-isoform fact, which is uncontroversial, and not for any potency ranking. Approval indications have changed since 2015 - belinostat and romidepsin's peripheral T-cell lymphoma approvals have since been affected by confirmatory-trial outcomes - so this citation should not be used for regulatory status.

    Evidence that argues against this (1)
    • Approval for haematological malignancies is not evidence in KMT2A-rearranged infant disease, and the paper cited is a functional-genetic dissection whose purpose was to test whether single-Hdac suppression phenocopies these compounds. Vorinostat's breadth is the reason its clinical effect cannot be assigned to any enzyme this beat discusses. Matthews GM et al. 2015
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    HDAC inhibitors including SAHA acutely release the positive transcription elongation factor b (P-TEFb) from its inhibitory 7SK snRNP complex, with HEXIM1 induction following as downstream feedback.

    Histone deacetylase inhibitors (HDACis) that release the positive transcription elongation factor b (P-TEFb) from its inhibitory complex also activate HIV transcription. · Bartholomeeusen K, Fujinaga K, Xiang Y, Peterlin BM · The Journal of Biological Chemistry 288(20):14400-14407 2013
    PMID 23539624DOI
    What this evidence cannot show

    The system is HIV latency in T-cell lines, not leukaemia, and no KMT2A-rearranged cell was studied. The claim carried here is about the biochemical action of the drug class on the 7SK reservoir, which this graph's P-TEFb beat treats as established for HMBA, SAHA and JQ1; whether that action matters in KMT2A-rearranged leukaemia has never been measured. The PMID, title and DOI were fetched from PubMed in this session and not copied from the sibling entry gene-hexim1, per this project's citation rule.

    Evidence that argues against this (1)
    • Every compound known to release P-TEFb produces a TRANSIENT rise followed by reassembly, and three independent negative feedbacks operate on the axis, so the sign of any effect depends on the timescale observed. A claim that vorinostat raises free P-TEFb in a patient over weeks of dosing does not follow from an acute release measured over hours. Bartholomeeusen K et al. 2012
Connected to