May 5, 2026 | Kyle Doherty
The investigational fourth-generation EGFR inhibitor HS-10504 demonstrated encouraging efficacy and a manageable safety profile at the 400-mg dose in patients with advanced EGFR-mutant non–small cell lung cancer (NSCLC) harboring the C797S resistance mutation, according to initial data from the phase 1 first-in-human HS-10504-101 trial (NCT06461156) that were presented at the 2026 AACR Annual Meeting.1
“HS-10504 monotherapy demonstrated an encouraging result in [patients with] advanced NSCLC harboring the C797S mutation post EGFR TKI treatment,” Zhanhong Xie, MD, lead study author and faculty within the Department of Respiratory and Critical Care Medicine and the Department of Oncology at The First Affiliated Hospital of Guangzhou Medical University in China, said in a presentation of the data. “Promising efficacy results [were seen] in a high unmet need population,” Xie added.
What makes HS-10504 a next-generation EGFR inhibitor?
Osimertinib’s (Tagrisso) role as a frontline standard of care for patients with EGFR-mutant advanced NSCLC has been upheld by data showing significantly improved progression-free survival (PFS) and overall survival (OS) vs first-generation EGFR TKIs. However, responses wane in the face of acquired resistance and development of progressive disease (PD).
The primary resistance mutation associated with the use of osimertinib in either the first or second-line setting is C797S, present in approximately 7% and 14% of cases, respectively. The mutation has also been documented with the use of other third-generation EGFR TKIs such as aumolertinib and furmonertinib.
HS-10504 is a novel, potent, fourth-generation EGFR TKI that is highly selective for EGFR sensitizing mutations and the C797S resistance mutation, with or without a concomitant T790M mutation.
How was the HS-10504-101 trial designed?
The single-arm, open-label, multicenter study enrolled patients with histologically or cytologically confirmed locally advanced or metastatic stage IIIB, IIIC, IV, or above NSCLC. An EGFR sensitizing mutation––either exon 19 deletion or L858R mutation––and a C797S mutation with or without a T790M mutation was required, alongside at least 1 target lesion according to RECIST 1.1 criteria and an ECOG performance status of 0 or 1.
In part 1 dose escalation a rolling 6 design was used to evaluate the following doses of HS-10504: 50 mg (n = 1), 100 mg (n = 4), 200 mg (n = 4), 300 mg (n = 3), 400 mg (n = 5), and 500 mg (n = 4).
In part 2 dose expansion two dose levels were evaluated: 300 mg (n = 31) and 400 mg (n = 30).
The primary end point was the maximum tolerated dose or maximum applicable dose.2 Secondary end points included safety, pharmacokinetics, objective response rate (ORR), disease control rate (DCR), duration of response (DOR), PFS, and OS.
Treatment was administered orally, once daily on a continuous basis until disease progression or other criteria for discontinuation were met.1
Data cutoff was December 25, 2025.
Who were the patients enrolled in this study?
Baseline characteristics were presented in the dose expansion population of patients who received HS-10504 at either 300 mg (n = 34) or 400 mg (n = 35). In the former population, the median age was 63.5 years (range, 37-78) and most patients were female (n = 19; 55.9%) and had stage IVB disease (n = 23; 67.6%) according to the American Joint Committee on Cancer (AJCC) 8th edition staging manual. ECOG performance status was largely 1 (n = 28; 82.4%) and most patients had a concomitant T790M mutation (n = 19; 55.9%). Central nervous system (CNS) metastases were found in the minority of patients (n = 6; 17.6%) as was prior exposure to chemotherapy (n = 10; 29.4%). With respect to prior treatment, patients had either received 1 (n = 13; 38.2%), 2 (n = 16; 47.1%), or 3 or more (n = 5; 14.7%) lines.
In the latter population the median age was 61.0 years (range, 35-78) and most patients were female (n = 25; 71.4%) and had stage IVB disease (n = 19; 54.3%) according to the AJCC 8th edition staging manual. ECOG performance status was largely 1 (n = 30; 85.7%) and most patients did not have a concomitant T790M mutation (n = 20; 57.1%). CNS metastases were found in the minority of patients (n = 6; 17.1%) as was prior exposure to chemotherapy (n = 14; 40.0%). With respect to prior treatment, patients had either received 1 (n = 13; 37.1%), 2 (n = 13; 37.1%), or 3 or more (n = 9; 25.7%) lines.
How safe and tolerable is HS-10504 at different dose levels?
The 400-mg dose was found to be manageable, Xie said, with most adverse effects (AEs) resolving after symptom management and/or dose interruption or modification. Only 1 patient discontinued treatment because of treatment-related AEs (TRAEs), supporting continuous dosing.
Median follow-up was 3.1 months (range, 0.1-11.5) and 5.3 months (range, 0.9-9.0) in the 300-mg and 400-mg cohorts, respectively. In the former population, any AEs, grade 3 or greater AEs, and any serious AEs occurred at rates of 91.2% (n = 31), 35.3% (n = 12), and 5.9% (n = 2), respectively. AEs led to treatment interruption in 20.6% (n = 7) of patients, and no events led to treatment reduction, discontinuation, or death. In the latter population, any AEs, grade 3 or greater AEs, and any serious AEs occurred at rates of 100.0% (n = 35), 65.7% (n = 23), and 31.4% (n = 11), respectively. AEs leading to treatment interruption, reduction, and discontinuation occurred at rates of 40.0% (n = 14), 42.9% (n = 15), and 2.9% (n = 1), respectively. No events led to death.
Common TRAEs that occurred in at least 25% of patients and grade 3 or greater TRAEs were also presented for the 400-mg dose cohort. Xie shared that most TRAEs were grade 1 or 2 and most grade 3 events that occurred in at least 10% of patients included decreased lymphocyte count (n = 11; 31.4%) and anemia (n = 5; 14.3%).
Any-grade and grade 3 or greater TRAEs, respectively, included increased blood bilirubin (82.9%; 5.7%), anemia (60.0%; 14.3%), increased alanine aminotransferase levels (57.1%; 2.9%), rash (54.3%; 0%), increased aspartate aminotransferase levels (48.6%; 0%), decreased lymphocyte count (45.7%; 31.4%), diarrhea (42.9%; 2.9%), decreased appetite (40.0%; 2.9%), increased blood creatinine (40.0%; 0%), hypoalbuminemia (37.1%; 0%), decreased white blood cell count (31.4%; 2.9%), hypokalemia (28.6%; 8.6%), increased blood urea (28.6%; 0%), increased conjugated bilirubin (28.6%; 2.9%), increased blood lactate dehydrogenase (25.7%; 0%), increased blood creatine phosphokinase (25.7%; 2.9%), and asthenia (25.7%; 0%).
What were the key response rates at 300 mg vs 400 mg?
Within the efficacy-evaluable population of the 300-mg cohort (n = 22), the ORR and confirmed ORR were each 31.8% (95% CI, 13.9%-54.9%), with 7 patients (31.8%) achieving a partial response (PR), 13 (59.1%) with stable disease (SD), and 2 (9.1%) with PD. The DCR was 90.9% (95% CI, 70.8%-98.9%) and the median time to response (TTR) was 1.5 months.
Within the efficacy-evaluable population of the 400-mg cohort (n = 34), the ORR and confirmed ORR were 50.0% (95% CI, 32.4%-67.6%) and 47.1% (95% CI, 29.8%-64.9%), respectively. No complete responses were observed; 17 patients (50.0%) achieved a PR, 14 (41.2%) had SD, and 3 (8.8%) experienced PD. The DCR was 91.2% (95% CI, 76.3%-98.1%), and the median TTR was 1.4 months.
At the latest data cutoff of March 6, 2026, among efficacy-evaluable patients in the 400-mg cohort, the ORR reached 52.9% (n = 18) and the confirmed ORR was 50.0% (n = 17).
These responses also proved not only higher but more durable in the 400-mg cohort. As of the updated data cutoff, the median PFS was 9.6 months (95% CI, 5.3-not available).
What does ctDNA clearance reveal about treatment activity?
Biomarker analyses further supported the clinical activity of HS-10504 at the 400-mg dose, with circulating tumor DNA (ctDNA) clearance data revealing rapid and deep molecular responses across key EGFR mutational profiles.
Among 20 patients with NSCLC who had detectable EGFR mutational profiles at baseline and were evaluable for changes in mutation allele levels on treatment, marked reductions in variant allele frequency were observed in both classical EGFR mutations and acquired resistance mutations after just one, 21-day cycle of treatment.
In patients harboring classical EGFR mutations (n = 20), 95.0% (n = 19) achieved at least 50% ctDNA clearance, and 70.0% (n = 14) achieved complete ctDNA clearance. Among those with the C797S resistance mutation (n = 13), 100% of patients achieved at least 50% clearance and 84.6% (n = 11) achieved complete clearance. In patients harboring the T790M resistance mutation (n = 9), 88.9% (n = 8) achieved at least 50% clearance, and 66.7% (n = 6) achieved complete clearance.
What comes next for HS-10504 in clinical development?
“These results suggest that HS-10504 is a potential best-in-class oral option, as a new targeted therapy with a promising efficacy and safety profile, for EGFR TKI–pretreated patients with a C797S mutation. Further clinical development is ongoing,” Xie concluded.
Disclosures: Xie had no financial relationships to disclose.
References
Xie Z, Wang Q, Wu L, et al. The safety, tolerability, and efficacy of HS-10504 in patients with EGFR mutation-positive locally advanced or metastatic NSCLC: initial results from a first-in-human phase 1 study. Cancer Res. 2026;86(suppl 8):CT302. doi:10.1158/1538-7445.AM2026-CT302
A study of HS-10504 in patients with advanced or metastatic non-small-cell lung cancer(NSCLC). ClinicalTrials.gov. Updated September 17, 2025. Accessed May 5, 2026. https://clinicaltrials.gov/study/NCT06461156
