QLS5132 led to a 50% ORR and manageable safety in PROC, supporting CLDN6 as a promising ADC target in early-phase clinical development.
Treatment with the novel Claudin 6 (CLDN6)–directed antibody-drug conjugate (ADC) QLS5132 led to encouraging activity not hindered by safety in patients with platinum-resistant ovarian cancer (PROC), according to data from a first-in-human phase 1 trial (NCT06932094) that were presented at the
Within the efficacy-evaluable population (n = 18), the objective response rate (ORR) was 50.0% (95% CI, 26.0%-74.0%), with 9 partial responses (PRs) occurring in the 3.2-mg/kg (n = 2), 4.8-mg/kg (n = 5), and 6.4-mg/kg (n = 2) cohorts. The overall disease control rate (DCR) was 94.4% (95% CI, 72.7%-99.9%).
“QLS5132 presented remarkable antitumor activity in patients with advanced PROC,” Tao Zhu, MD, presenting study author and chief physician and vice president of Zhejiang Cancer Hospital in Hangzhou, China, said. “QLS5132 was well tolerated, especially under potential recommended phase 2 dose [RP2D] levels with a wider therapeutic window.”
Why study CLDN6 as a therapeutic target?
CLDN6 is integral to the formation and polarity of epithelial cell junctions within intercellular tight junctions and is involved in the differentiation of stem cells into epithelial cells. CLDN6 expression is also highly expressed in solid tumors such as ovarian and lung cancer, portending poor prognosis, with scant expression in healthy tissue.
QLS5132 is a highly selective ADC targeting CLDN6 that has shown an expanded therapeutic window and robust efficacy in PROC in preclinical research, according to Zhu.
How was the trial designed to evaluate the antitumor activity and safety of QLS5132?
Eligible patients were at least 18 years old and had histologically or cytologically confirmed advanced solid tumors that had failed or had no standard therapy available for use, with an ECOG performance status of 0 or 1.2
In dose escalation, which employed an accelerated titration design with a Bayesian optimal interval, patients received 1.6 mg/kg (n = 1), 3.2 mg/kg (n = 4), 4.8 mg/kg (n = 12), 5.6 mg/kg (n = 8), or 6.4 mg/kg (n = 3) of intravenous QLS5132 once every 3 weeks (Q3W).1
In dose expansion, patients with PROC, which comprised cohort 1, were randomly assigned to receive either 4.8 mg/kg or 5.6 mg/kg of QLS5132. Cohorts 2, 3, 4, and 5 enrolled patients with non–small cell lung cancer (NSCLC); gastric/gastroesophageal junction tumors; other solid tumors; and solid tumors with low CLDN6 expression or 0 expression on immunohistochemical (IHC) staining.
As of January 21, 2026, 28 patients with PROC had been enrolled into cohort 1. Median follow-up was 2.2 months (IQR, 0.9-3.7), and 22 patients (78.6%) were receiving ongoing therapy.
The median age in the PROC cohort was 57.5 years (range, 49-76). Most patients had an ECOG performance status of 1 (n = 23; 82.1%), had ovarian cancer (n = 26; 92.9%) as opposed to fallopian tube cancer (n = 2; 7.1%), and had received prior bevacizumab (Avastin; n = 25; 89.3%) and PARP inhibition (n = 22; 78.6%). Approximately half of patients (n = 12; 42.9%) were CLDN6 positive by IHC. The median number of prior lines of therapy was 3 (range, 1-6).
Were the responses seen with QLS5132 durable?
Response was upheld for over 3 months in 5 of the 9 patients who achieved PR at the 3.2-mg/kg and 4.8-mg/kg dose level.
“Patients [treated] at the 3.2-mg/kg dose level maintained PR for over 6 months, suggesting that patients at a higher dose level may yield more durable antitumor activity,” Zhu said.
When responses were further evaluated according to dose level, an ORR of 55.6% (95% CI, 21.2%-86.3%) was seen in the 4.8-mg/kg cohort (n = 9), which included 5 PRs and a DCR of 100% (95% CI, 66.4%-100%). At a dose level of 3.2 mg/kg or above, an ORR of 52.9% (95% CI, 27.8%-77.0%) was seen, with 9 PRs and a DCR of 100% (95% CI, 80.5%-100%).
“Patients with low or undetectable CLDN6 protein expression still showed clinical response,” Zhu said. Although 2 PRs were achieved by patients with undetectable CLDN6 expression, the limited sample size prevented a definitive conclusion from being drawn.
The median relative dose intensity across all dose levels was 99.8% (range, 48.5%-102.5%).
What was the reported safety profile of the agent?
With respect to safety, 1 dose-limiting toxicity of grade 4 platelet count decrease occurred at the 6.4-mg/kg dose level, which has since recovered to baseline level. The maximum-tolerated dose was not reached, and no treatment-related adverse effects (TRAEs) led to death or treatment discontinuation.
Across the dose levels (n = 28), TRAEs occurred in 92.9% (n = 26) of patients, 32.1% (n = 9) of which were grade 3 or greater. Serious TRAEs occurred in 1 (3.6%) patient, and TRAEs leading to dose delay or dose reduction occurred in 14.3% (n = 4) and 10.7% (n = 3) of patients, respectively.
“Common TRAEs were gastrointestinal and hematological AEs, mostly grade 1 or 2,” Zhu said.
Any-grade TRAEs that occurred in at least 30% of patients included nausea (85.7%), anorexia (60.7%), weakness (53.6%), anemia (46.5%; grade ≥3, 7.1%), decreased platelet count (35.7%; grade ≥3, 14.3%), decreased neutrophil count (39.3%; grade ≥3, 10.7%), decreased white blood cell count (42.9%; grade ≥3, 7.1%), vomiting (39.3%), increased aspartate aminotransferase level (39.3%), and increased gamma-glutamyl transferase level (32.1%).
Interstitial lung disease, ocular toxicity, and febrile neutropenia did not occur, and no more than 1 case each of alopecia, bone pain, back pain, hypoesthesia, and oral mucositis occurred.
What did the pharmacokinetic assessment reveal about the agent?
Pharmacokinetic exposure increased in a dose-proportional manner over the dose range. The mean elimination half-life surpassed 5.5 days, supporting Q3W dosing.
Moreover, the mean molar ratio of released payload to ADC area under the curve was less than 0.2%, suggesting limited off-target release.
As far as the RP2D was concerned, investigators observed that doses of 4.8 mg/kg or greater had a Ctrough that could cover the efficacious exposure in the mouse PA-1 model.
Enrollment into the PROC, NSCLC, gastric cancer, and other cancer cohorts is ongoing.
Disclosures: Zhu disclosed being an employee of Zhejiang Cancer Hospital.
References
