Immunotherapy Comes of Age in Prostate Cancer:

The Clues Hidden in Failed Prostate Cancer Trials

IPCSG Newsletter  •  Treatment Advances

What the “Failures” Are Finally Teaching Us

A patient-friendly review of the science — Summer 2026

BLUF (Bottom Line Up Front): 

 For fifteen years, the checkpoint-inhibitor drugs that revolutionized melanoma, lung, and kidney cancer failed one prostate-cancer trial after another — six large Phase 3 studies came back negative. But hidden inside those “failures” was a small group of men who had dramatic, lasting remissions. By studying those rare responders one cell at a time, researchers have now identified why most prostate tumors resist immunotherapy (immune-suppressing “SPP1” macrophages) and what lets a few tumors melt away (unusual killer T-cell populations). Today, only men whose tumors carry specific genetic signatures — mismatch-repair deficiency (dMMR), high microsatellite instability (MSI-H), or high tumor mutational burden (TMB) — have an FDA-approved immunotherapy option (pembrolizumab). But a genuinely new generation of treatments is now in trials: bispecific T-cell engagers (xaluritamig, pasritamig, VIR-5500) that physically drag immune cells onto tumor cells, plus radioligand-plus-immunotherapy combinations and tumor-reprogramming nanoparticles. Experts on both coasts now say the field has reached a real turning point. If you have advanced prostate cancer, the single most useful step is to make sure your tumor has had comprehensive genomic and mismatch-repair testing — it is the key that unlocks both today's approved drug and tomorrow's trials.

Why Prostate Cancer Was Immunotherapy's Great Disappointment

Prostate cancer was supposed to be an immunotherapy success story. In April 2010, the FDA approved sipuleucel-T (Provenge) — the first therapeutic cancer vaccine of any kind — after the pivotal IMPACT trial showed it extended median survival by about 4 months (25.8 versus 21.7 months) in men with metastatic castration-resistant prostate cancer (mCRPC). The benefit was modest, and curiously the vaccine rarely lowered PSA or shrank tumors on scans, but it proved a landmark point: the immune system could be turned against prostate cancer.

Then the checkpoint-inhibitor era arrived. Drugs that release the immune system's brakes — the PD-1 and CTLA-4 inhibitors — transformed treatment for melanoma, lung cancer, kidney cancer, bladder cancer, and head-and-neck cancer. Prostate cancer was expected to follow. It did not. One large Phase 3 trial after another came back negative — six in all. Prostate tumors were labeled immunologically “cold”: they contained few functioning tumor-fighting T cells and showed little natural anti-tumor activity, so releasing the brakes on an immune response that was barely running produced nothing. As University of Minnesota prostate-cancer specialist Emmanuel Antonarakis put it, early optimism was quickly replaced by skepticism. Many considered prostate cancer immunotherapy's biggest letdown.

But the averages hid something important. In virtually every failed trial, a handful of men had striking, durable responses — some lasting years. As one oncologist noted, doctors tend to remember those rare outliers, not the hundreds who did not benefit. The trials had simply enrolled “all comers” and the exceptional responders were statistically diluted out. Instead of concluding that immunotherapy just doesn't work in prostate cancer, researchers began asking a sharper, more productive question: why do a rare few respond so completely, while most do not?

The Exceptional Responders: Clues From the Men Who Did Beat the Odds

The best-characterized responders share a genetic fingerprint. Roughly 2–4% of prostate cancers are mismatch-repair deficient (dMMR), meaning the cell's DNA proofreading machinery is broken. These tumors accumulate enormous numbers of mutations (a high tumor mutational burden, or TMB) and often show high microsatellite instability (MSI-H). All that genetic chaos makes the tumor look profoundly “foreign” to the immune system — exactly the trait that predicts response to checkpoint drugs. This is the biology behind pembrolizumab's approval (discussed below), but it doesn't explain how the immune system actually eradicates such a tumor.

To answer that, Antonarakis and first author Alexander Tsai reported a remarkable case (published in Cell Reports Medicine, 2026). A 67-year-old man arrived with an 8-cm prostate mass invading the bladder and rectum — the highest possible grade, Gleason 5+5=10 — that was mismatch-repair deficient with an ultra-high mutation burden roughly ten times the median. Hormone therapy failed within three months. But once he started pembrolizumab, his PSA fell to undetectable within two cycles; by four cycles, scans showed no residual tumor; and after his prostate was removed, the pathologist found no cancer at all. Deep sequencing showed that the thousands of mutations present before treatment had all vanished — a complete molecular response. In the researchers' assessment, the patient appears to be cured.

When the team profiled his immune cells before and after treatment, two unusual populations stood out: “NK-like” T cells (killer T cells that had borrowed the weaponry of natural-killer cells) and “double-positive” T cells carrying both CD4 and CD8 markers — a combination normally seen only in immature immune cells. A single NK-like T-cell clone multiplied nearly 50-fold right after the first dose and stayed expanded. Crucially, when the team looked back at other patients, these same two populations expanded in immunotherapy responders but not in non-responders. In other words, these odd T cells may be a signature — and eventually a target — for the response everyone wants to reproduce.

Decoding the “Cold” Tumor: The SPP1 Macrophage Problem

If the case report explains how a rare tumor is destroyed, the other half of the puzzle is why most tumors shut immunotherapy down. Lawrence Fong's laboratory (now at Fred Hutch / University of Washington) tackled this with single-cell profiling, sequencing nearly 150,000 individual cells from tumor biopsies spanning three stages of disease and mapping 14 distinct myeloid (immune) cell subsets. Published in Nature in late 2024, the atlas delivered a surprise: the team expected the key difference to lie in T cells, but it lay in the macrophages.

A specific macrophage population — marked by high levels of a gene called SPP1 — became more and more dominant as the cancer advanced, peaking in castration-resistant disease. These SPP1-high macrophages are powerfully immune-suppressing: they shut down T-cell activity and drive immune exhaustion. And they carried a detail that explains a decade of drug failures — they barely express CSF1R, the very receptor that an entire class of “myeloid-targeting” drugs had been built to attack. The field had been aiming at the wrong doorknob.

The atlas also pointed to a better target. Instead of relying on CSF1R, these macrophages lean on adenosine signaling — they thrive in low-oxygen tumor regions and pump out enzymes that flood their surroundings with adenosine, a molecule that quiets T cells. When Fong's team combined an adenosine-blocking drug with a checkpoint inhibitor, roughly 20–30% of men with mCRPC responded — and, tellingly, the responders were the ones whose tumors were rich in SPP1 macrophages. For the first time, a biomarker was pointing to who might benefit.

The New Frontier: Dragging the Immune System Onto the Tumor

The most clinically advanced new approach doesn't wait for a cold tumor to attract T cells on its own. Bispecific T-cell engagers are engineered antibodies with two grips: one hand grabs a protein on the prostate-cancer cell, the other hand grabs the CD3 receptor on a passing T cell, physically clamping the two together so the T cell kills the tumor cell. Antonarakis has argued that the future of prostate immunotherapy is not PD-1 or CTLA-4 inhibition but these CD3-targeting engagers. Three are leading the pack, each aimed at a different tumor protein:

Drug (target)	Early-phase results	Status
Xaluritamig (STEAP1 × CD3) Amgen	Phase 1 in heavily pre-treated mCRPC: ~49% had a ≥50% PSA drop and ~24% had tumor shrinkage overall — higher at target doses (about 59% PSA response, 41% shrinkage). Cytokine release syndrome was common but manageable.	Advanced to Phase 3 (XALience, NCT07213674)
Pasritamig (KLK2 × CD3) Johnson & Johnson	First-in-human Phase 1: ~42% had a ≥50% PSA drop in the recommended-dose group. Notably well tolerated (fewer than 10% had cytokine release) and given as an outpatient infusion — a practical advantage.	Phase 3 initiated Sept 2025 (NCT07164443)
VIR-5500 (PSMA × CD3) Vir Biotechnology	Phase 1 (data cut Jan 2026): at the highest dose levels, 82% had a ≥50% PSA drop (14 of 17 evaluable), 53% had a ≥90% drop, and 45% had measurable tumor shrinkage. A “dual-masked” design aims to keep the drug switched off until it reaches the tumor, limiting side effects.	Phase 3 expected to begin in 2027 (Phase 1 NCT05997615)

As Fong has said, given how clearly active these drugs are in the clinic, it would be surprising if at least one didn't produce a positive Phase 3 result. A word of caution for patients reading the percentages: these are early-phase numbers in small groups of very advanced patients, and a PSA response is not the same as proven longer survival — that is exactly what the Phase 3 trials are now testing.

Radiation That Wakes Up the Immune System

A second strategy pairs immunotherapy with radioligand therapy — the PSMA-targeted radiation (such as ¹⁷⁷Lu-PSMA-617 / Pluvicto) already familiar to many IPCSG members. When radiation kills tumor cells, the dying cells spill out inflammatory signals that can rouse a sleeping immune system. In the Australian PRINCE trial (published in Lancet Oncology, 2026), the combination of ¹⁷⁷Lu-PSMA-617 plus pembrolizumab produced a ≥50% PSA decline in about 76% of patients, with deep and durable responses in some. Because the trial was small and single-arm (no comparison group), we can't yet separate the radiation's effect from the immunotherapy's — but the pattern of “break the suppression first, then immunotherapy works better” keeps recurring. Related combinations, such as radioligand therapy paired with a PARP inhibitor (the LuPARP study), are being explored on the same logic.

Nanoparticles That Turn “Cold” Tumors “Hot”

The earliest-stage idea is the most futuristic. Michelle Bradbury's team at Weill Cornell has repurposed ultrasmall silica nanoparticles — called Cornell Prime dots (C′ dots), originally built for medical imaging — to home in on prostate tumors via a PSMA-targeting tag. Reported in Cancer Research in June 2026, the particles do two things at once in mice: they push tumor cells toward self-destruction, and they reprogram the immune landscape from “cold” to “hot.” Combined with checkpoint blockade, the treatment produced complete or near-complete remissions in about 4 of 10 mice; adding a macrophage-targeting (CSF-1R) drug pushed that to 5 of 10. This is still animal research — a first-in-human trial design is being developed — but it directly attacks the SPP1-macrophage problem Fong's lab identified.

What This Means for You Today

What is actually available now — and what isn't

Available (FDA-approved):

• Pembrolizumab (Keytruda) — approved on a “tissue-agnostic” basis for tumors that are MSI-H/dMMR (first approved 2017, converted to full approval and reaffirmed for genitourinary cancers in 2025–2026) or that have high tumor mutational burden (≥10 mutations/megabase; 2020). Only a minority of prostate cancers qualify — which is exactly why genomic testing matters.
• Sipuleucel-T (Provenge) — the cell-based vaccine, approved for asymptomatic/minimally symptomatic mCRPC.

Experimental (clinical trials only): the bispecific engagers (xaluritamig, pasritamig, VIR-5500), radioligand-plus-immunotherapy combinations, adenosine-blocking combinations, and the Cornell-dot nanoparticles. The bispecifics are still several years from possible approval; the nanoparticle work is earlier still.

The single most actionable takeaway is about testing. Whether a man can access today's approved immunotherapy — or qualify for tomorrow's trials — hinges on knowing his tumor's biology. Ask whether your tumor has had comprehensive genomic profiling and specific testing for mismatch-repair status (dMMR), microsatellite instability (MSI-H), and tumor mutational burden (TMB). These can be assessed on tumor tissue and, increasingly, through a blood-based “liquid biopsy.” A man who turns out to be MSI-H or dMMR is in a fundamentally different situation than the “average” prostate-cancer patient the failed trials described.

There is also an honest, humbling frontier the researchers themselves emphasize: a subset of men respond dramatically without any known biomarker — no mismatch-repair defect, no obvious explanation. Antonarakis compares high mutational burden to buying many lottery tickets: more tickets improve your odds, but a single lucky ticket can still win. Finding those hidden winners — the men with a highly immune-responsive tumor but an ordinary-looking genomic report — is the question the field has not yet cracked.

Questions you might bring to your oncologist

• Has my tumor been tested for mismatch-repair deficiency (dMMR), microsatellite instability (MSI-H), and tumor mutational burden (TMB)? If not, can it be — on tissue or by liquid biopsy?
• Given my results, am I a candidate for pembrolizumab, or is it not indicated for my tumor type?
• Are there clinical trials of bispecific T-cell engagers (for example, STEAP1-, KLK2-, or PSMA-targeted) or radioligand-plus-immunotherapy combinations that I might qualify for?
• If I've already had PSMA radioligand therapy, does that affect my eligibility for these newer approaches?
• What does a “PSA response” in these early trials actually mean for my expected survival and quality of life?

After two decades of watching immunotherapy fall short, the researchers who lived through those failures are, in their own words, hopeful again — not because the checkpoint drugs finally worked, but because studying why they didn't has pointed the way to treatments built for how prostate cancer actually behaves. For patients, the practical message is steadier and simpler: know your tumor's biology, keep asking about trials, and understand that the ground is genuinely shifting.

Verified Sources

1. Brzostowicki D. “The Clues Hidden in Failed Prostate Cancer Immunotherapy Trials.” Medscape, 2026. (Originating review that prompted this article.) https://www.medscape.com/
2. Lyu A, Fong L, et al. “Evolution of myeloid-mediated immunotherapy resistance in prostate cancer.” Nature, 2024 (SPP1-high macrophages, low CSF1R, adenosine dependence). https://www.nature.com/articles/s41586-024-08290-3
3. Tsai AK, Lozada JR, … Antonarakis ES. “Case of complete response to immunotherapy in MMR-deficient prostate cancer associated with NK-like and CD4+CD8+ T cells.” Cell Reports Medicine, 2026. https://www.cell.com/cell-reports-medicine/fulltext/S2666-3791(26)00306-X
4. Kelly WK, Danila DC, Lin C-C, et al. “Xaluritamig, a STEAP1 × CD3 XmAb 2+1 Immune Therapy for Metastatic Castration-Resistant Prostate Cancer: Results from Dose Exploration in a First-in-Human Study.” Cancer Discovery, 2024;14:76–89; with ESMO 2024 dose-expansion update. https://www.urotoday.com/conference-highlights/esmo-2024/esmo-2024-prostate-cancer/154804-esmo-2024-xaluritamig-a-steap1-x-cd3-xmab-2-1-immune-therapy-in-patients-pts-with-metastatic-castration-resistant-prostate-cancer-mcrpc-initial-results-from-dose-expansion-cohorts-in-a-phase-1-study.html
5. Baldini C, et al. “Phase 1 study results of JNJ-78278343 (pasritamig) in mCRPC.” Journal of Clinical Oncology, 2025;43:5017 (ASCO 2025). https://ascopubs.org/doi/10.1200/JCO.2025.43.16_suppl.5017
6. Johnson & Johnson. “Johnson & Johnson unveils first-in-human results for pasritamig, showing early anti-tumor activity in prostate cancer.” Press release, June 1, 2025. https://www.jnj.com/media-center/press-releases/johnson-johnson-unveils-first-in-human-results-for-pasritamig-showing-early-anti-tumor-activity-in-prostate-cancer
7. Vir Biotechnology. “Vir Biotechnology Reports Positive Updated Phase 1 Results for PSMA-targeting, PRO-XTEN Dual-masked T-Cell Engager VIR-5500 in Patients with Metastatic Prostate Cancer.” Press release, Feb 23, 2026. https://www.businesswire.com/news/home/20260223892941/en/
8. Sandhu S, Hofman MS, et al. “[177Lu]Lu-PSMA-617 in combination with pembrolizumab for treatment of metastatic castration-resistant prostate cancer (PRINCE): a single-arm, phase 1b/2 study.” The Lancet Oncology, 2026. https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(26)00017-3/fulltext
9. Bradbury M, Wiesner U, et al. Cornell Prime dots (C′ dots) reprogram the prostate tumor microenvironment. Cancer Research, June 15, 2026 (Weill Cornell Medicine / Cornell newsroom summaries). https://news.weill.cornell.edu/news/2026/06/experimental-treatment-directly-kills-prostate-tumor-cells-while-reawakening-antitumor  |  https://www.eurekalert.org/news-releases/1132001
10. Kantoff PW, Higano CS, Small EJ, et al. “Sipuleucel-T immunotherapy for castration-resistant prostate cancer (IMPACT).” N Engl J Med, 2010;363:411–422. https://doi.org/10.1056/NEJMoa1001294  |  Review: Cheever MA, Higano CS, Clin Cancer Res, 2011;17:3520–3526. https://aacrjournals.org/clincancerres/article/17/11/3520/12151/
11. U.S. Food and Drug Administration / Merck. Pembrolizumab tissue-agnostic approvals: MSI-H/dMMR (2017; full approval reaffirmed for solid tumors including genitourinary cancers, 2025–2026) and TMB-high ≥10 mut/Mb (2020). https://www.urologytimes.com/view/fda-grants-full-approval-to-pembrolizumab-for-msi-h-dmmr-solid-tumors-including-genitourinary-cancers  |  FDA approval summary: https://pmc.ncbi.nlm.nih.gov/articles/PMC8416776/
12. Altomare NJ, Hussain M, VanderWeele DJ, et al. “Response to pembrolizumab in advanced prostate cancer with predictive biomarkers.” The Oncologist, 2025;30(3):oyaf025 (PSA responses concentrated in MSI-H patients). https://doi.org/10.1093/oncolo/oyaf025

This article is provided by the Informed Prostate Cancer Support Group for educational purposes and reflects published research as of Summer 2026. It is not medical advice and does not replace consultation with your own physicians. Early-phase clinical-trial results describe small groups of patients and may change as larger studies mature; PSA responses do not by themselves prove a survival benefit. Treatment decisions should be made with your care team based on your individual diagnosis, genomic testing, prior treatments, and overall health.