Gastric Cancer Drug Zolbetuximab Drives Widespread Albumin Loss in Real-World Patients

A new real-world study from Aichi Cancer Center Hospital in Japan has revealed that hypoalbuminemia—dangerously low levels of the blood protein albumin—is far more common during treatment with the gastric cancer drug zolbetuximab than clinical trials ever suggested. In the retrospective analysis, published in Cancer Reports, every single one of the 39 patients evaluated developed some degree of albumin decline, and more than half experienced severe hypoalbuminemia, a finding that stands in stark contrast to the modest rates reported in the pivotal phase III trials that led to the drug’s approval.

Zolbetuximab is a chimeric monoclonal antibody that targets claudin-18 isoform 2, or CLDN18.2, a protein found on gastric cancer cells but also on the normal mucosal lining of the stomach. Approved in combination with platinum- and fluoropyrimidine-based chemotherapy as a first-line treatment for patients with CLDN18.2-positive advanced gastric cancer, the drug has a distinctive safety profile dominated by nausea and vomiting. But as the new study makes clear, another adverse event—hypoalbuminemia and the peripheral edema it causes—has increasingly been recognized as a characteristic and potentially treatment-derailing complication of zolbetuximab-based therapy.

The discrepancy between trial data and clinical reality is striking. In an integrated analysis of the SPOTLIGHT and GLOW phase III trials, hypoalbuminemia was reported in 19.3 percent of patients for all grades combined, and in only 3.8 percent for grade 3 or higher events. In the Aichi cohort, by comparison, all 39 patients developed hypoalbuminemia of any grade, 24 patients—61.5 percent—reached grade 2, and seven patients, or 17.9 percent, reached grade 3. Serum albumin levels declined in all but one patient, with the median time to reach the lowest level during treatment being 34 days. Peripheral edema of any grade occurred in 46.1 percent of patients, and diuretics were administered to 41 percent.

The researchers, led by Taro Mizuno and Yukiya Narita, conducted their single-center retrospective study among patients with advanced gastric cancer who initiated zolbetuximab plus chemotherapy between July 2024 and August 2025 and received at least two cycles of treatment. Eligibility required immunohistochemical confirmation of HER2-negative, CLDN18.2-positive gastric or esophagogastric junction adenocarcinoma. The median age of the cohort was 61 years, most patients had a gastric primary tumor, and the peritoneum was the most common metastatic site, present in 84.6 percent of cases. Ascites—fluid accumulation in the abdomen—was present in 64.1 percent of patients, and 79.5 percent had diffuse-type adenocarcinoma, a histological subtype known for aggressive behavior.

To capture clinically meaningful albumin deterioration, the team defined severe hypoalbuminemia as a minimum serum albumin level below 2.0 grams per deciliter—the threshold for grade 3 events under the Common Terminology Criteria for Adverse Events—or a decrease of at least 1.0 grams per deciliter from baseline to the nadir during treatment. The second criterion was adopted because absolute grading alone can miss significant deterioration in patients who start with relatively high albumin levels. Using this definition, 20 patients, or 51.3 percent, developed severe hypoalbuminemia.

The search for risk factors produced two clinically important findings. Patients who experienced nausea, vomiting, or both persisting from the completion of the first treatment cycle to the initiation of the second had a significantly higher incidence of severe hypoalbuminemia than those without these symptoms—63.0 percent versus 25.0 percent. More striking still, no cases of severe hypoalbuminemia were observed among the five patients who had undergone prior total gastrectomy, compared with 58.8 percent among those whose stomach remained intact. Nausea, vomiting, or abdominal pain severe enough to require interruption of the first zolbetuximab infusion, and anorexia during the first cycle, showed trends toward association but did not reach statistical significance.

The gastrectomy finding carries particular mechanistic weight. Because zolbetuximab targets CLDN18.2, which is expressed not only on cancer cells but also on the normal gastric mucosa, the drug may damage the stomach lining itself, causing a protein-losing enteropathy—a syndrome in which plasma proteins leak through the gastrointestinal wall. This hypothesis is supported by an emerging case report describing zolbetuximab-induced protein-losing enteropathy characterized by diffuse gastric mucosal redness and edema, mucous cell detachment and fundic gland disruption on biopsy, and direct visualization of albumin leakage on technetium-99m albumin scintigraphy. The absence of severe hypoalbuminemia in patients without a stomach strongly suggests that CLDN18.2-expressing normal gastric mucosa is central to susceptibility to the drug’s albumin-depleting effects.

Perhaps the most actionable finding concerns reversibility. Recovery of serum albumin from the nadir back to baseline was observed in 18 patients, or 46.2 percent, with a median time to recovery of 64.5 days. No patients received albumin supplementation. Critically, patients who temporarily interrupted treatment for any reason after reaching their albumin nadir were significantly more likely to recover than those who continued treatment—66.7 percent versus 28.6 percent. Treatment was interrupted in 18 patients, with both zolbetuximab and chemotherapy withheld in 16 cases and zolbetuximab alone in two, for a median duration of 14 days. This pattern suggests that zolbetuximab-induced mucosal injury may be at least partially reversible when the drug is withheld, though the study could not determine whether such recovery translates into clinical benefit, nor whether edema improvement was attributable to interruption or to the diuretics most edema patients received.

Importantly, the presence of severe hypoalbuminemia did not appear to compromise treatment efficacy. Among the 10 patients with measurable lesions assessed by RECIST version 1.1 criteria, the objective response rate was 20.0 percent in patients without severe hypoalbuminemia and 60.0 percent in those with it, a difference that was not statistically significant given the small numbers. Among the 25 patients with baseline ascites, ascites response rates were 50.0 percent and 40.0 percent respectively. Survival analyses, with a median follow-up of 7.1 months, showed no significant differences in progression-free or overall survival according to severe hypoalbuminemia in either first-line or later-line settings, though the authors caution that these exploratory analyses were limited by wide confidence intervals, immature data, and potential time-dependent bias.

The study’s authors conclude that severe hypoalbuminemia is common during zolbetuximab plus chemotherapy in routine clinical practice, and that the absence of prior total gastrectomy and the presence of persistent post-infusion nausea or vomiting may help identify patients at increased risk of profound albumin decline. They also note that all patients in the cohort received acid-suppressive agents and gastric mucosal protectants under an institutional protocol, yet hypoalbuminemia still occurred universally—suggesting these measures alone are insufficient for prevention, and that careful monitoring with timely treatment modification may be a more appropriate strategy than excessive prophylactic medication. Limitations include the small sample size, the retrospective single-institution design, the possibility that disease progression contributed to albumin decline, and the lack of routine endoscopic assessment to directly correlate clinical findings with mucosal injury. Further validation, the team emphasizes, is needed to determine the optimal management of zolbetuximab-associated hypoalbuminemia without compromising the treatment’s anticancer efficacy.

Subject of Research: Zolbetuximab-associated hypoalbuminemia in CLDN18.2-positive advanced gastric cancer

Article Title: Retrospective Analysis of Zolbetuximab‐Associated Hypoalbuminemia in Claudin 18.2‐Positive Advanced Gastric Cancer

Article References: Mizuno, T., Narita, Y., Ando, T., Ishizuka, Y., Umegaki, S., Honda, K., Masuishi, T., Taniguchi, H., Kadowaki, S., Ando, M., Tajika, M., & Muro, K. (2026). Retrospective Analysis of Zolbetuximab‐Associated Hypoalbuminemia in Claudin 18.2‐Positive Advanced Gastric Cancer. Cancer Reports, 9(10), Article e70709. https://doi.org/10.1002/cnr2.70709

Image Credits: AI Generated

DOI: 10.1002/cnr2.70709

Keywords: zolbetuximab, gastric cancer, hypoalbuminemia, CLDN18.2, protein-losing enteropathy, peripheral edema, monoclonal antibody, advanced gastric cancer, treatment interruption, albumin recovery, chemotherapy, real-world study