Stripping Cancer's Invisibility Cloak: What the Latest Immune Evasion and Chemo-Alternative Research Actually Promises — and What It Doesn't

At the 2025 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago — the world's largest cancer conference — researchers presented a cluster of therapies aimed at two of oncology's most stubborn problems: tumors that hide from the immune system, and patients who cannot tolerate or no longer respond to chemotherapy. The results ranged from early-phase signals to Phase 3 head-to-head victories over standard-of-care chemo. But the distance between a conference podium and a patient's bedside remains vast, shaped by biology, money, and structural inequity.

How Tumors Go Invisible — and the New Drugs Designed to Strip That Cover

Cancer cells survive in part by manipulating the molecular signals that tell the immune system what to attack. Several distinct mechanisms were addressed by therapies presented at ASCO 2025.

ERAP1 Inhibition: Rewriting the Tumor's Molecular ID

GRWD5769, developed by Greywolf Therapeutics, is a first-in-class oral inhibitor of endoplasmic reticulum aminopeptidase 1 (ERAP1) — an enzyme that trims peptides before they are displayed on MHC class I molecules on the cell surface [1]. MHC-I molecules act as molecular ID badges: they present fragments of internal proteins to T cells, which then decide whether the cell is healthy or cancerous. Tumors can co-opt ERAP1 to distort this display, effectively hiding tumor-specific antigens from immune surveillance.

By inhibiting ERAP1 on a cyclical three-weeks-on, three-weeks-off schedule, GRWD5769 generates alternating antigenic repertoires — broadening the T-cell response and preventing the T-cell exhaustion that results from chronic exposure to the same antigens [1]. In a trial of 83 patients with cervical, bladder, liver, bowel, lung, and head-and-neck cancers — all of whom had failed prior treatment including immunotherapy — GRWD5769 combined with cemiplimab (an anti-PD-1 antibody) produced tumor shrinkage of at least 30% in 15 patients [2]. Disease was halted for at least six months in 18% of cervical cancer patients, 32% of liver cancer patients, 36% of bladder cancer patients, 38% of head-and-neck cancer patients, 51% of bowel cancer patients, and 55% of lung cancer patients [2]. Objective response rates ranged from 13% to 36% depending on tumor type [1].

These are early-phase numbers, and the trial population — patients who had exhausted other options — makes the results more notable than a raw percentage might suggest.

Fascin Inhibition: Re-sensitizing Tumors to Checkpoint Therapy

NP-G2-044, developed by Novita Pharmaceuticals, takes a different approach. It is a small-molecule inhibitor of fascin, an actin-bundling protein that promotes metastasis and contributes to the immunosuppressive tumor microenvironment [3]. In a Phase 2 trial of 45 patients with solid tumors that had progressed on prior anti-PD-(L)1 therapy, NP-G2-044 at 1,600 mg daily combined with anti-PD-1 therapy achieved a disease control rate of 76% and an objective response rate of 21%, including four complete responses [3]. Novita plans to begin a pivotal Phase 3 study in platinum-resistant ovarian cancer.

CD47: A Cautionary Tale

Not every immune evasion target has translated smoothly to the clinic. CD47 — the so-called "don't eat me" signal that cancer cells overexpress to avoid being engulfed by macrophages — has been a major research focus for over a decade [4]. Despite strong preclinical rationale, multiple CD47-targeting programs have been discontinued after disappointing clinical results [4]. The challenge has been that CD47 is expressed on healthy cells too, particularly red blood cells, leading to dose-limiting anemia. This history underscores that identifying a mechanism of immune evasion and successfully drugging it are separate problems.

Source: OpenAlex

Data as of Jan 1, 2026CSV

Beating Chemo at Its Own Game: Head-to-Head Survival Data

Tarlatamab in Small Cell Lung Cancer

The most striking head-to-head comparison came from the Phase 3 DeLLphi-304 trial, which randomized 509 patients with previously treated small cell lung cancer (SCLC) to receive either tarlatamab — a bispecific T-cell engager targeting delta-like ligand 3 (DLL3) — or standard chemotherapy (topotecan, lurbinectedin, or amrubicin) [5].

At a median follow-up of 11.2 months, tarlatamab produced a median overall survival of 13.6 months versus 8.3 months for chemotherapy — a 5.3-month advantage [5]. Six-month survival rates were 76% for tarlatamab versus 62% for chemo; at 12 months, the gap widened to 53% versus 37% [5]. Tarlatamab also produced fewer serious treatment-related adverse events and better patient-reported outcomes on cough and dyspnea [6]. The FDA granted tarlatamab (marketed as Imdelltra by Amgen) full approval for extensive-stage SCLC in November 2025 [7].

Source: ASCO 2025 / DeLLphi-304 Phase 3 Trial

Data as of Jun 1, 2025CSV

Chemo-Sparing Approaches in NSCLC

In non-small cell lung cancer (NSCLC), the KRYSTAL-7 study evaluated adagrasib plus pembrolizumab — without traditional chemotherapy — as first-line therapy for patients with KRAS G12C-mutated tumors, reporting an overall response rate of 44.3% and median overall survival of 18.3 months [8]. Meanwhile, the CheckMate-816 study's long-term data showed that combining neoadjuvant nivolumab with chemotherapy before surgery reduced the risk of recurrence and death by nearly 40% in resectable NSCLC, with benefits sustained beyond three years [8].

China-based Akeso's ivonescimab, a first-in-class bispecific antibody targeting both PD-1 and VEGF, demonstrated a median overall survival of 27.9 months versus 23.7 months for PD-1 inhibitor plus chemotherapy in advanced squamous NSCLC (hazard ratio 0.66) — results selected for the 2026 ASCO Plenary Session, a first for a China-originated oncology drug [9].

The Long Road from Conference to Clinic

Conference presentations generate optimism, but the regulatory pipeline is unforgiving. Historically, only about 5.3% of oncology drugs that enter Phase 1 trials ultimately receive FDA approval [10]. The steepest drop-off occurs between Phase 2 and Phase 3, where roughly 71% of candidates fail, most often due to insufficient efficacy (52% of failures) or safety concerns (24%) [10]. The median time from first-in-human study to approval for an oncology drug is 13.1 years [10].

Source: BIO/QLS Advisors Clinical Development Success Rates 2011-2020

Data as of Jan 1, 2025CSV

GRWD5769 and NP-G2-044 are both in Phase 2 — a stage where approximately 29% of oncology drugs advance to Phase 3 [10]. Tarlatamab, having completed Phase 3 and received FDA approval, is the exception, not the rule.

Cost Projections

Novel biologic therapies routinely exceed $10,000 per month in the United States — more than double the median monthly household income [11]. Immunotherapy costs for a full treatment course frequently surpass $100,000 per patient [12]. Biosimilar competition has begun to reduce prices for older biologics: Q1 2025 average selling price discounts reached 52% for trastuzumab, 49% for bevacizumab, and 66% for rituximab [11]. But the checkpoint inhibitors central to these new combination regimens — pembrolizumab (Keytruda) and nivolumab (Opdivo) — do not face biosimilar competition until their patents expire between 2028 and 2033 [11].

Who Gets Left Behind

Clinical trial enrollment in oncology does not mirror the patient population. According to ASCO's 2025 State of Cancer Care report, Black patients represent 10% of cancer prevalence but only 6% of therapeutic trial participants; Hispanic patients represent 7% of prevalence but only 3% of enrollment [13]. Patients who are female, non-White, or at age extremes are consistently less likely to participate [13]. Head-and-neck cancer trials, specifically, were found not to represent the diversity of the patient population — a trend that did not improve after implementation of the Affordable Care Act [14].

Source: ASCO State of Cancer Care 2025

Data as of Jun 1, 2025CSV

These gaps matter because treatment effects can vary by genetics, comorbidity burden, and social determinants of health. When the evidence base is built on a non-representative population, the resulting treatment guidelines may not perform equally for everyone.

Geographically, over 60% of immunotherapy trials are concentrated in North America and Western Europe, with East Asia (particularly China) emerging as a third hub [12]. Fewer than 3% of global immunotherapy trials include sites in Sub-Saharan Africa or lower-middle-income Southeast Asian nations [12]. Cold-chain logistics, manufacturing capacity, and intellectual property law remain structural barriers to access. The EU's Health Technology Assessment Regulation, which took effect in January 2025, may accelerate regulatory timelines across the bloc but does not address affordability directly [12].

The Case for Chemotherapy

For all the momentum behind targeted and immune-based therapies, chemotherapy remains the backbone of cancer treatment in several contexts. In many cancer types — including pancreatic, ovarian, and triple-negative breast cancer — chemotherapy still produces the strongest evidence for survival benefit, particularly in resource-limited settings where immunotherapy infrastructure (cold-chain storage, infusion centers, biomarker testing) does not exist [15].

Chemotherapy's mechanism of action — broadly cytotoxic rather than targeted — also means it works regardless of specific tumor mutations or immune markers. For patients whose tumors lack druggable targets or who cannot access biomarker testing, chemotherapy is often the only viable option. In the neoadjuvant setting, chemotherapy combined with immunotherapy has shown better outcomes than immunotherapy alone, as demonstrated in CheckMate-816 — suggesting that chemo's role is evolving rather than disappearing [8].

Long-term safety data for many newer agents remain limited. Tarlatamab's follow-up is 11.2 months; GRWD5769 and NP-G2-044 are even earlier in development. Chemotherapy's toxicity profile, while harsh, is thoroughly characterized over decades. Researchers who frame new therapies as replacements for chemo before multi-year survival and safety data mature risk creating expectations that outrun the evidence.

Resistance: The Tumor Fights Back

Immune evasion is not a one-time event. Tumors can re-acquire resistance through multiple pathways: downregulating MHC-I expression, secreting immunosuppressive cytokines, recruiting regulatory T cells, or exploiting metabolic changes in the tumor microenvironment [15]. A 2025 study also revealed that cancer-induced nerve injury within tumors promotes immunotherapy resistance — a mechanism that was not previously recognized [15].

The fascin inhibitor NP-G2-044 and the ERAP1 inhibitor GRWD5769 were both designed with resistance in mind: GRWD5769's cyclical dosing schedule is specifically intended to prevent T-cell exhaustion [1], while NP-G2-044 targets the metastatic microenvironment that enables immune escape [3]. Combination strategies — pairing checkpoint inhibitors with targeted therapies, cytokine variants (IL-2, IL-18), or personalized cancer vaccines — are the dominant research framework for addressing acquired resistance [15].

How frequently resistance emerges after initial response remains an open question in the early-phase trials. Longer follow-up data will be critical.

Follow the Money

The pharmaceutical industry's role in oncology research is substantial but shifting. The share of oncology trial sponsorship by EU- and US-based pharma companies fell from 59% in 2014 to 28% in 2023, with biotechnology firms and regional innovation hubs — particularly in China — filling the gap [16]. At ASCO 2025, the conference specifically highlighted the role of federal sponsorship in trial design, examining alternatives to industry-funded models [16].

Industry sponsorship introduces well-documented biases. A meta-analysis in oncology trial literature has consistently shown that industry-funded studies report favorable efficacy outcomes more frequently than independently funded studies — not necessarily through data fabrication, but through selective endpoint choice, comparator selection, and publication timing [16]. The tarlatamab DeLLphi-304 trial was sponsored by Amgen; the GRWD5769 trial by Greywolf Therapeutics; NP-G2-044 by Novita Pharmaceuticals. ASCO requires disclosure of conflicts of interest, and the Open Payments database provides public records of industry payments to physicians, but the structural incentive to present favorable results at the world's highest-profile oncology conference remains [16].

Federal- and nonprofit-sponsored trials — like the cooperative group model used in CheckMate-816 — tend to select more clinically meaningful comparators and endpoints. ASCO's 2025 report specifically called for expanding this model [13].

What This Means

The therapies presented at ASCO 2025 represent genuine scientific advances. GRWD5769 addresses a previously undrugged mechanism of immune evasion. NP-G2-044 offers a potential rescue strategy for patients who have failed checkpoint therapy. Tarlatamab has already delivered a survival advantage over chemotherapy significant enough to earn FDA approval.

But the pathway from these results to broad patient benefit is constrained by biology (resistance), economics (pricing), regulation (approval timelines), and equity (who participates in trials and who can afford the drugs). Academic publication on cancer immunotherapy and immune evasion has grown from 427 papers in 2011 to over 14,600 in 2025 [17]. The volume of research is not the bottleneck. The bottleneck is translating that research into treatments that reach the patients who need them — across demographics, geographies, and income levels — before the next wave of resistance takes hold.