romidepsin
FK228, FR901228, depsipeptide, NSC 630176
Romidepsin is a bicyclic depsipeptide natural product from the bacterium Chromobacterium violaceum and a potent inhibitor of class I histone deacetylases. It is a genuine prodrug: a disulfide bridge is reduced inside the cell to unmask the thiol that chelates zinc in the enzyme's active site. It holds a US licence for cutaneous T-cell lymphoma, but its confirmatory randomised trial in peripheral T-cell lymphoma failed, that indication was withdrawn, and European regulators refused authorisation altogether.
Mechanism
Romidepsin is a redox-activated prodrug. The intact molecule is a 16-membered macrocycle containing D-valine, D-cysteine, (Z)-dehydrobutyrine and L-valine, closed through an ester linkage (hence depsipeptide) to a (3S,4E)-3-hydroxy-7-mercapto-4-heptenoic acid unit, with an intramolecular disulfide bridge spanning the ring. Once inside the cell, glutathione and thioredoxin reduce that disulfide, releasing a free butenyl thiol. That thiol is the business end: it inserts into the narrow hydrophobic channel of the histone deacetylase catalytic pocket and chelates the active-site Zn²⁺ ion, blocking the deacetylation reaction.
Potency is concentrated on class I enzymes (HDAC1 and HDAC2 are inhibited at low nanomolar concentrations, HDAC3 somewhat less, HDAC8 weakly), with much lower activity against class II enzymes. The consequence is hyperacetylation of histones H3 and H4, relaxation of chromatin, and re-expression of genes silenced during malignant transformation, including cell cycle inhibitors such as p21^WAF1/CIP1. Non-histone substrates matter too: acetylation of HSP90 disables its chaperone function and destabilises client oncoproteins, and acetylation of p53 and α-tubulin alters their behaviour. In T-cell lymphomas the net effects include G1 and G2/M arrest, reactive oxygen species accumulation, sensitisation to death-receptor signalling and induction of apoptosis, alongside effects on the tumour microenvironment. Romidepsin is also a substrate for P-glycoprotein, which contributes to resistance.
What the research shows
Two independent single-arm phase 2 studies established the cutaneous T-cell lymphoma indication. The NCI-led multi-institutional trial reported by Piekarz and colleagues in 71 patients found an overall response rate of 34% with 6 complete responses and a median duration of response of 13.7 months, while the international pivotal study reported by Whittaker and colleagues in 96 patients with refractory disease found a 34% response rate, 6% complete responses and a median response duration of 15 months, with clinically meaningful reduction in pruritus. In relapsed or refractory peripheral T-cell lymphoma, the pivotal study reported by Coiffier and colleagues in 130 patients found a 25% overall response rate with 15% complete or unconfirmed complete responses and a median duration of response of 28 months, which was long enough to appear striking in a disease with dismal outcomes. None of these studies had a control arm.
The randomised evidence then went the other way. The Ro-CHOP phase 3 trial randomised 421 patients with previously untreated peripheral T-cell lymphoma to romidepsin plus CHOP chemotherapy or CHOP alone. Median progression-free survival was 12.0 versus 10.2 months (HR 0.81, p = 0.096). The primary endpoint was not met, while grade 3 or worse thrombocytopenia (50% versus 10%), neutropenia, anaemia and leucopenia were all substantially more common in the romidepsin arm. The final analysis published in 2024, with longer follow-up, confirmed no progression-free or overall survival benefit. The company withdrew the US peripheral T-cell lymphoma indication in 2021. Combination work continues. Romidepsin with oral azacitidine or with pralatrexate has shown activity in angioimmunoblastic and other T-follicular-helper-phenotype lymphomas, where the epigenetic rationale is strongest given the frequency of TET2, DNMT3A and IDH2 mutations, but none of this has yet produced randomised confirmation.
Evidence assessment
Mixed evidence
Romidepsin holds an FDA licence for cutaneous T-cell lymphoma, but that approval rests entirely on two single-arm phase 2 studies with overall response rates of 34% and complete response rates of only around 6%. No randomised trial has ever shown a survival or progression benefit. The one large randomised phase 3 trial, Ro-CHOP in previously untreated peripheral T-cell lymphoma (n = 421), was negative on its primary endpoint and added toxicity, and its 2024 final analysis confirmed no benefit; the peripheral T-cell lymphoma indication was voluntarily withdrawn in the US in 2021. The EMA refused a marketing authorisation and confirmed that refusal on re-examination, so the drug is not licensed in the EU or UK at all. All six cited studies were verified against PubMed. Calling this evidence base strong would misrepresent it.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Phase II multi-institutional trial of the histone deacetylase inhibitor romidepsin as monotherapy for patients with cutaneous T-cell lymphoma Preclinical only
Overall response rate 34% with 6 complete responses; median duration of response 13.7 months. One of the two studies supporting FDA accelerated approval in CTCL.
Final results from a multicenter, international, pivotal study of romidepsin in refractory cutaneous T-cell lymphoma Preclinical only
Overall response rate 34% with 6% complete responses and median response duration 15 months; 43% of patients with significant baseline pruritus achieved clinically meaningful relief. No control arm.
Results from a pivotal, open-label, phase II study of romidepsin in relapsed or refractory peripheral T-cell lymphoma after prior systemic therapy Preclinical only
Overall response rate 25% with 15% complete or unconfirmed complete responses; median duration of response 28 months. Supported the accelerated approval that was later withdrawn.
Romidepsin plus CHOP versus CHOP in patients with previously untreated peripheral T-cell lymphoma: results of the Ro-CHOP phase III study (conducted by LYSA) Preclinical only
Negative trial. Median progression-free survival 12.0 versus 10.2 months (HR 0.81, p = 0.096), primary endpoint not met, with substantially greater haematological toxicity in the romidepsin arm. Led to withdrawal of the US PTCL indication.
Romidepsin plus cyclophosphamide, doxorubicin, vincristine, and prednisone versus cyclophosphamide, doxorubicin, vincristine, and prednisone in patients with previously untreated peripheral T-cell lymphoma: final analysis of the Ro-CHOP trial Preclinical only
Extended follow-up confirmed no progression-free or overall survival advantage from adding romidepsin to CHOP in first-line peripheral T-cell lymphoma.
Safety
Nausea and vomiting are very common and warrant routine antiemetic cover; fatigue, anorexia and dysgeusia are frequent. Haematological toxicity (thrombocytopenia, neutropenia, anaemia and lymphopenia) is dose-limiting in many patients and was markedly worse when romidepsin was added to CHOP. Infections, including serious and fatal ones, occur, and Epstein-Barr virus and hepatitis B reactivation have been reported. Cardiac effects require specific attention: ST-segment and T-wave changes on ECG are common, QTc prolongation occurs, and electrolytes (particularly potassium and magnesium) must be repleted before and during treatment; caution applies in congenital long QT syndrome, significant cardiac disease and with other QT-prolonging drugs. Tumour lysis syndrome has been reported in patients with high tumour burden. As a CYP3A4 and P-glycoprotein substrate, exposure is altered by strong inhibitors such as azole antifungals and by inducers such as rifampicin. Romidepsin is embryo-foetal toxic.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Not licensed. The EMA's Committee for Medicinal Products for Human Use adopted a negative opinion on 19 July 2012 for the peripheral T-cell lymphoma indication, confirmed the refusal on re-examination on 15 November 2012, and formal refusal of marketing authorisation followed on 12 February 2013 (dates verified against the EMA record). The decisive reason was that the single-arm pivotal study included no comparator, so the committee could not judge clinical benefit; a good-manufacturing-practice certification gap at the production site was raised initially but was resolved during re-examination. There is consequently no EU or UK marketing authorisation, and any use would be through unlicensed import or clinical trial routes. |
| United States | FDA accelerated approval November 2009 for cutaneous T-cell lymphoma in patients who have received at least one prior systemic therapy. This indication remains in force. A separate accelerated approval for relapsed or refractory peripheral T-cell lymphoma was granted in June 2011 and voluntarily withdrawn in 2021 after the confirmatory Ro-CHOP trial failed to verify clinical benefit. Intravenous, specialist use. |
| WADA (sport) | Not on the WADA Prohibited List. Romidepsin is a cytotoxic epigenetic agent with no anabolic, hormonal or masking properties. |
Questions
Because the molecule as administered cannot inhibit anything. Its disulfide bridge has to be reduced by intracellular glutathione and thioredoxin to release a free thiol, and it is that thiol that reaches into the histone deacetylase pocket and grabs the catalytic zinc. It is an unusually neat example of redox activation being built into a natural product.
The EMA refused it. The CHMP issued a negative opinion on 19 July 2012, confirmed it on re-examination on 15 November 2012, and formal refusal followed on 12 February 2013. The decisive reason was that the pivotal study had no comparator arm, so the committee could not judge clinical benefit against existing treatments. A manufacturing good-practice certification problem was also raised, but that one was resolved during re-examination.
Yes, in the United States in 2021. Accelerated approval in 2011 came with an obligation to confirm benefit in a randomised trial. That trial, Ro-CHOP, missed its primary endpoint of progression-free survival and added considerable haematological toxicity, so the indication was voluntarily withdrawn. The cutaneous T-cell lymphoma indication remains.
A microbe. It was isolated from Chromobacterium violaceum, a Gram-negative bacterium, during screening of fermentation broths in the early 1990s. It was originally noted for its ability to reverse the transformed phenotype of ras-transformed cells, before its histone deacetylase target was identified.
ECG changes, particularly ST-segment and T-wave abnormalities, are common, and QTc prolongation occurs. Potassium and magnesium are repleted before and during treatment, and caution applies with other QT-prolonging drugs, congenital long QT syndrome and significant cardiac disease.