IMA901 and multi-peptide vaccines

IMA901; related multi-peptide constructs IMA950 (glioma), IMA910 (colorectal), GAPVAC (actively personalised glioma vaccine), PolyPEPI, Vx-001; "TUMAP" = tumour-associated peptide

Oncology & imaging

Awaiting medical reviewEN ES DE FR PT ZH+ Follow for updates

Note on this profile. These are investigational cancer vaccines given only inside clinical trials under oncology supervision. None is available outside trials, and nothing here describes self-preparation.

Summary

Class: Fixed cocktails of multiple synthetic tumour-associated peptides — IMA901 contains ten peptides (nine HLA-A*02:01-restricted class I, one class II) identified from HLA ligandome analysis of human kidney tumours — given with GM-CSF, usually after a single low dose of cyclophosphamide
Mechanism: Vaccinates against many tumour antigens at once so that losing any single antigen does not allow the tumour to escape, with cyclophosphamide pre-treatment intended to deplete regulatory T cells first
Routes studied: Intradermal injection with GM-CSF at the same site; frequent dosing over the first weeks then maintenance. Cyclophosphamide 300 mg/m² intravenously three days before the first vaccination
Half-life: Not meaningful; peptides are cleared within hours, immune memory is the durable element
Storage: Refrigerated trial product, prepared by pharmacy
Regulatory status: Not approved anywhere; the lead programme was discontinued after a negative phase 3. See region table

Doses are as reported in studies or by the community — never recommendations.

1What it is

IMA901 was the most rigorously developed multi-peptide cancer vaccine of its generation and is worth a profile because its failure settled a question. Rather than guessing at antigens, its developers used mass spectrometry to read out which peptides are actually presented on HLA molecules by human renal cell carcinoma tissue and not by normal kidney — the "HLA ligandome" — and assembled the ten most promising into a single fixed vaccine. A single low dose of cyclophosphamide was given before the first vaccination, based on earlier observations that this depletes regulatory T cells and improved outcomes in a phase 2.

The broader class is the same idea in different diseases: IMA950 in glioma, Vx-001 (a telomerase peptide construct) in NSCLC, PolyPEPI in colorectal cancer, and GAPVAC, which went a step further and assembled each patient's vaccine from a pre-manufactured warehouse of peptides matched to their own tumour and HLA type.

2How it's thought to work

Ten peptides covering multiple tumour antigens are injected intradermally with GM-CSF, which recruits dendritic cells.

Several antigens are targeted simultaneously, so a tumour cannot escape by silencing one.

The class II peptide is included to recruit CD4+ helper T cells, which short class I peptides alone recruit poorly.

Cyclophosphamide given three days earlier is intended to transiently deplete regulatory T cells and lift suppression of the vaccine response.

Sunitinib, the standard-of-care backbone in the phase 3, was expected to be compatible or even synergistic through its own effects on myeloid-derived suppressor cells.

A fixed set of ten tumour peptides for renal cell carcinoma, given with GM-CSF. Phase III did not improve survival, which is part of why the field moved to individualised antigens.Schematic — not to scale

3What the research shows

Graded A (negative)

Advanced renal cell carcinoma

IMPRINT, a randomised open-label phase 3 trial in 339 patients with treatment-naive metastatic renal cell carcinoma, compared sunitinib plus IMA901 (with cyclophosphamide pre-treatment) against sunitinib alone. There was no improvement in overall survival — median overall survival was roughly 33 months in both arms, with a hazard ratio slightly favouring the control arm — and no improvement in progression-free survival or response rate (Rini 2016, Lancet Oncology). The trial was unambiguously negative and the programme was discontinued. [1]

The phase 2 that led to it

A randomised phase 2 had reported that patients mounting T-cell responses to multiple IMA901 peptides survived longer, and that cyclophosphamide pre-treatment was associated with better outcomes (Walter 2012, Nature Medicine). Both findings were biomarker-based and landmark-biased: immune response was measured in patients well enough to survive long enough to be measured. This is the single most instructive feature of the IMA901 story, because the phase 2 analysis was sophisticated, biologically well argued, and still did not predict the randomised result. [2]

Graded A/B

Immunogenicity

The vaccine reliably did what it was designed to do immunologically: multi-peptide T-cell responses were induced in a substantial fraction of patients in both phase 2 and phase 3. Immunogenicity without clinical benefit is the recurring verdict across this whole class.

IMA950 in glioma

Phase 1/2 studies of the 11-peptide glioma vaccine, with poly-ICLC as adjuvant in later iterations, reported immune responses and acceptable safety in small cohorts. No randomised efficacy result.

GAPVAC and warehouse-personalised vaccines

The GAPVAC-101 phase 1 study in newly diagnosed glioblastoma showed that a per-patient vaccine assembled from a peptide warehouse was feasible and immunogenic (Hilf 2019, Nature). Feasibility and immunogenicity only; small numbers. [3]

Graded A (negative)

Vx-001 in NSCLC

A randomised phase 2b of this telomerase-peptide construct as maintenance in advanced NSCLC did not improve overall survival. Consistent with the class.

4Reported side effects and interactions

Literature-reported

injection-site reactions in the large majority, largely from GM-CSF. Flu-like symptoms, fatigue, low-grade fever. In IMPRINT the serious-event profile was dominated by sunitinib (hypertension, fatigue, hand–foot syndrome, cytopenias) and adding the vaccine did not meaningfully worsen it. Cyclophosphamide at 300 mg/m² as a single dose contributes transient cytopenias and nausea. One notable finding: the vaccine arm had a numerically higher rate of some adverse outcomes, and the hazard ratio did not favour it — a reminder that "well tolerated" is not the same as "harmless".

Community-reported

Community use: essentially none documented.

Theoretical. several IMA901 antigens are expressed at low levels in normal tissue, so autoimmunity was a predicted risk; it did not emerge at the observed response magnitudes.

Interactions. sunitinib (studied, no benefit from the combination); cyclophosphamide (deliberate pre-treatment); GM-CSF (adjuvant); corticosteroids and other immunosuppressants would be expected to blunt responses.

Markers researchers commonly monitor

From trial protocols:

Marker	Why it's watched
Multi-peptide T-cell response breadth and magnitude (ELISpot, intracellular cytokine staining)	the designed readout. Responses were induced; they did not predict benefit in the randomised setting. The single most important caveat to carry forward from this programme.
*HLA-A02:01 genotype**	eligibility gate.
Regulatory T-cell (CD4+CD25highFoxP3+) frequency before and after cyclophosphamide	the pharmacodynamic test of whether the pre-treatment did what it was supposed to.
Myeloid-derived suppressor cell subsets	explored in the phase 2 biomarker work as a predictor of non-response.
Overall survival, progression-free survival, RECIST imaging	the endpoints.
CBC after cyclophosphamide; blood pressure, thyroid function and CMP on sunitinib	safety monitoring driven by the backbone therapy, not the vaccine.

Free for everyone. Safety information is never behind a paywall.

5Formulation and administration context

Supplied as refrigerated peptide solutions prepared by trial pharmacy and injected intradermally, with GM-CSF at the same site, on a front-loaded schedule — in IMPRINT, up to 17 injections across nine months. Cyclophosphamide 300 mg/m² was infused three days before the first vaccination. HLA-A*02:01 typing was an eligibility requirement for IMA901. There is no self-prepared, compounded or outpatient form and no dose calculation appropriate to publish.

6Commonly studied alongside

Sunitinib; cyclophosphamide as a regulatory-T-cell-depleting pre-treatment; GM-CSF and poly-ICLC as adjuvants; checkpoint inhibitors in later-generation designs. Conceptually related: gp100 peptide vaccine (profile 231), Nelipepimut-S (profile 232), GV1001 (profile 233), neoantigen peptide vaccines (profile 238) — which is where the field went after this class failed.

Informational cross-links. Not stack recommendations.

7Regulatory status by region

Region	Status (checked —)
United States	Not FDA-approved; investigational. The IMA901 programme was discontinued after IMPRINT
European Union	Not authorised
United Kingdom	Not licensed
Australia	Not approved
South Africa	Not registered with SAHPRA
Sport	Not listed by WADA; no performance application

8Open questions

Is a fixed cocktail of shared tumour-associated peptides fundamentally too weak, or was renal cell carcinoma plus sunitinib the wrong test?
Why do immune-response biomarkers in single-arm and phase 2 cancer-vaccine studies so consistently fail to predict randomised outcomes?
Does the warehouse-personalised approach (GAPVAC) or true neoantigen vaccination (profile 238) solve the problem, or inherit it?

9References

1Rini BI et al. (2016). IMA901, a multipeptide cancer vaccine, plus sunitinib versus sunitinib alone, as first-line therapy for advanced or metastatic renal cell carcinoma (IMPRINT): a multicentre, open-label, randomised, controlled, phase 3 trial. Lancet Oncology. PubMed
2Walter S et al. (2012). Multipeptide immune response to cancer vaccine IMA901 after single-dose cyclophosphamide associates with longer patient survival. Nature Medicine. PubMed · DOI
3Hilf N et al. (2019). Actively personalized vaccination trial for newly diagnosed glioblastoma (GAPVAC-101). Nature. PubMed · DOI
4Rampling R et al. (2016). A cancer research UK first time in human phase I trial of IMA950 (novel multipeptide therapeutic vaccine) in patients with newly diagnosed glioblastoma. Clinical Cancer Research. PubMed · DOI
5Weinschenk T et al. (2002). Integrated functional genomics approach for the design of patient-individual antitumor vaccines. Cancer Research. PubMed
6Immatics (2016). IMPRINT phase 3 results — sponsor announcement. PubMed

References are being resolved to PubMed/DOI identifiers; links search PubMed for the citation until then.

10Community

forum threads →

community protocols include this →

Threads and protocols are visible to members. Joining is free.

Self-reported side effects Phase 2

Not enough data yet. 0 of the 50 members needed have logged this compound for 4+ weeks. We don't show anything below 50, so no one can be identified.

11Page history

31 Oct 2026Profile published — awaiting medical review

Licensed CC BY-NC-SA 4.0. Found an error? Report it · Editorial standards — corrections are logged here.