Nature Medicine Reports Phase 2 Data from ASPEN-06
A four-drug combination built around the investigational agent evorpacept has been carried through the phase 2 portion of a randomized phase 2/3 study in advanced gastric cancer, according to a report published online in Nature Medicine on 24 September 2026. The paper, issued under DOI 10.1038/s41591-026-04700-3, centers on the ASPEN-06 trial, which enrolled patients with HER2-positive advanced gastric cancer.
The regimen under investigation pairs evorpacept with three agents already familiar in this disease setting: trastuzumab, ramucirumab and paclitaxel. Because the trial randomizes participants, its design is intended to distinguish the contribution of the novel agent from the activity of the established backbone, rather than crediting the combination as a single undifferentiated treatment.
The Regimen, Component by Component
Understanding what makes this trial notable requires unpacking the four agents involved. Each targets a different layer of tumor biology, and the logic of the study rests on how those layers are expected to interact.
Trastuzumab: the HER2 anchor
Trastuzumab is a monoclonal antibody directed at HER2, the receptor tyrosine kinase that defines this molecular subset of gastric cancer. Its role is to bind the receptor on the surface of tumor cells, blocking proliferative signaling and flagging those cells for recognition by immune effector mechanisms. It is the foundational targeted therapy for HER2-positive gastroesophageal disease.
Ramucirumab: angiogenesis blockade
Ramucirumab targets VEGFR-2, the receptor through which vascular endothelial growth factor signaling drives the formation of new blood vessels feeding a tumor. By interfering with that pathway, the antibody aims to restrict vascular support for tumor growth and is used in advanced gastric and gastroesophageal junction cancers.
Paclitaxel: the cytotoxic partner
Paclitaxel is a taxane chemotherapy that stabilizes microtubules and disrupts the process of cell division, producing cytotoxic pressure on rapidly proliferating cells. It is a common chemotherapy companion in advanced gastric cancer and is routinely deployed alongside antibody-based therapy.
Evorpacept: the investigational agent
Evorpacept is designed to block CD47, a cell-surface protein that transmits a so-called don't-eat-me signal to macrophages. By interrupting that signal, the agent is intended to allow innate immune cells to engulf tumor cells — a mechanism that, in principle, could sharpen the effect of antibody-based targeting. The ASPEN-06 trial places that hypothesis directly alongside a standard antibody-and-chemotherapy backbone.
Why Pair an Innate Immune Checkpoint With HER2 Blockade
The scientific rationale behind the combination connects macrophage biology to antibody therapy. When an antibody such as trastuzumab coats a tumor cell, immune cells can recognize that coating and act on it. CD47 blockade is theorized to remove a brake on that process, making it easier for macrophages to complete the engulfment step. Layering a CD47-directed agent onto an anti-HER2 antibody, an anti-angiogenic antibody and a taxane therefore represents an attempt to attack the tumor through signaling blockade, vascular restriction, direct cytotoxicity and innate immune clearance at the same time.
That kind of stacking carries its own questions. More mechanisms mean more potential for overlapping toxicity, and the sequencing and dosing of four agents in a population of patients with advanced disease is a practical challenge. A randomized comparison is the structure best suited to answering whether the added agent contributes anything beyond the backbone.
What a Phase 2 Portion Inside a Phase 2/3 Trial Means
Seamless trial designs that embed a randomized phase 2 portion within a larger phase 2/3 program have become common in oncology, particularly for combinations where investigators want an early, controlled signal before committing to a large registrational study. Randomization at the phase 2 stage matters: single-arm data in advanced gastric cancer are difficult to interpret because the disease course varies considerably between patients and because historical benchmarks drift as supportive care and later lines of therapy change.
An embedded phase 2 also allows the trial to evaluate endpoints that read out relatively quickly, such as objective response and progression-free survival, while preserving the option to expand into a phase 3 comparison powered for definitive outcomes. The structure effectively compresses what would otherwise be two separate studies into one continuous program.
Reading a Mid-Stage Oncology Signal With Care
Phase 2 results in this disease area should be interpreted with an eye on their limits. Response rates can look encouraging without translating into a durable survival advantage, and progression-free survival gains do not always carry over to overall survival. Cross-trial comparisons are especially unreliable when patient populations, prior treatment exposure and HER2 expression definitions differ.
It is also worth remembering that the phase 2 portion of a phase 2/3 trial is designed to inform a decision about the next stage, not to settle the clinical question. The value of the ASPEN-06 readout lies in whether it supports continuing the program and how it shapes the design of any later comparison.
Where This Fits in Gastric Cancer Research
HER2-positive gastric cancer has become a testing ground for layered antibody strategies, with researchers repeatedly asking whether adding another mechanism to an established backbone can improve on what trastuzumab, ramucirumab and chemotherapy already achieve. Studies such as ASPEN-06 belong to that line of inquiry, and their results feed into a broader conversation about how many mechanisms can usefully be combined before tolerability becomes the limiting factor.
Key points at a glance
- The ASPEN-06 trial is a randomized phase 2/3 study; the Nature Medicine paper covers its phase 2 portion.
- The population is patients with HER2-positive advanced gastric cancer.
- The regimen combines evorpacept with trastuzumab, ramucirumab and paclitaxel.
- Evorpacept is an investigational CD47-directed agent intended to support innate immune clearance of tumor cells.
- Randomization is intended to isolate the experimental agent's contribution from the standard backbone.
- The report was published online on 24 September 2026 under DOI 10.1038/s41591-026-04700-3.
What Comes Next
The path forward depends on whether the phase 2 data justify advancing the regimen into the phase 3 component of the program, and on which endpoints regulators and investigators consider most persuasive in this setting. For now, the ASPEN-06 publication adds a documented data point to the ongoing effort to determine how innate immune checkpoint blockade might be integrated into HER2-directed treatment of advanced gastric cancer — a question that will only be resolved by the controlled, later-stage evidence that seamless trial designs are built to generate.
This article is based on reporting by Nature Medicine. Read the original article.
Originally published on nature.com
