
Immunotherapy combined with chemotherapy has become the standard first-line treatment for driver gene-negative non-small cell lung cancer (NSCLC). However, owing to the heterogeneity of the tumor immune microenvironment and complex mechanisms of primary and acquired resistance, the efficacy of existing regimens has reached a plateau, with only limited improvements in patient survival. How to optimize first-line maintenance strategies and overcome treatment resistance has therefore become a key clinical challenge. Therapeutic cancer vaccines offer a promising approach to addressing this unmet need by stimulating specific antitumor immune responses.
At the 2026 European Society for Medical Oncology (ESMO) Congress, important data were presented from an investigator-initiated Phase Ib/IIa study led by Professor Jianchun Duan’s team at the National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences, evaluating LK101, a personalized neoantigen dendritic cell (DC) vaccine, in combination with tislelizumab, with or without pemetrexed, as first-line maintenance treatment for patients with locally advanced or metastatic NSCLC. The study aimed to assess the efficacy and safety of introducing personalized neoantigen vaccine maintenance therapy in patients whose disease had not progressed following first-line chemoimmunotherapy induction. During the congress, Oncology Frontier invited Professor Duan for an exclusive interview to discuss the rationale behind the study design, its clinical findings, and the development strategies for therapeutic cancer vaccines.
Oncology Frontier: Could you introduce the background of this study and explain why you chose to focus on the first-line maintenance setting? What were the main considerations underlying the study design?
Professor Jianchun Duan: Lung cancer has now fully entered the era of precision diagnosis and treatment. For patients with advanced NSCLC harboring driver gene alterations, targeted therapies have substantially prolonged progression-free survival (PFS) and overall survival (OS). However, although immunotherapy combined with chemotherapy has further improved survival compared with conventional chemotherapy in patients without driver gene alterations, the gains in median PFS and median OS remain limited, and substantial unmet clinical needs persist.
To address this clinical bottleneck, we sought to identify new opportunities for therapeutic intervention from a mechanistic perspective. We know that tumor neoantigen vaccines can specifically expand neoantigen-reactive T cells and may act synergistically with immune checkpoint inhibitors to enhance antitumor activity. Based on this rationale, we developed the central hypothesis of this study: introducing the personalized neoantigen dendritic cell (DC) vaccine LK101 during first-line maintenance treatment could further activate patients’ endogenous antitumor immune surveillance through this combination strategy, potentially prolonging first-line PFS and ultimately translating into an OS benefit.
Regarding the study design, LK101 is an autologous DC vaccine that uses messenger RNA (mRNA) encoding up to 30 patient-specific tumor neoantigens to achieve precision targeting. Its production involves a highly customized, closed-loop process. First, whole-exome sequencing (WES) and RNA sequencing are performed on the patient’s tumor tissue to predict and select neoantigens. Peripheral blood leukocytes are then collected, monocytes are enriched, and these cells are induced to differentiate into mature DCs. Finally, mRNA encoding the selected neoantigens is loaded into the DCs through electroporation, and the resulting cells are reinfused into the patient.
The study enrolled patients with advanced NSCLC who had no driver gene alterations, a programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) of at least 1%, and no disease progression following four to six cycles of induction therapy with platinum-based doublet chemotherapy combined with an anti-PD-1 monoclonal antibody.
Patients received intradermal LK101 injections in combination with tislelizumab, with or without pemetrexed. The dosing schedule consisted of an induction phase of four weekly administrations, a boost phase of three administrations every three weeks, and a maintenance phase with administration every six weeks. Treatment continued for up to one year.
Through this early-phase study, we hoped to preliminarily determine whether the regimen could achieve the expected improvement in PFS with an acceptable safety profile, while also exploring whether these benefits could translate into longer-term survival gains.
Oncology Frontier: Considering the design characteristics of LK101, could you discuss its efficacy and safety as maintenance treatment for patients with locally advanced or metastatic NSCLC whose disease has not progressed following first-line chemoimmunotherapy? How do you assess its clinical potential in this setting and the patient populations that may benefit?
Professor Jianchun Duan: As of March 31, 2026, a total of 26 patients had been enrolled in the study, of whom 25 received LK101 treatment. Among the 21 patients evaluable for efficacy, four achieved a confirmed partial response (PR), corresponding to an overall objective response rate (ORR) of 19.0% and a disease control rate (DCR) of 95.2%.
Notably, outcomes differed between histological subtypes. The ORR was 50% among patients with squamous cell carcinoma, compared with 6.7% among those with non-squamous disease.
Although the median follow-up is currently only 9.5 months, the survival curves have been encouraging. Calculated from the initiation of first-line treatment, the nine-month and 12-month PFS rates were 85.4% and 70.4%, respectively.
Even more encouraging were the immunological findings. Enzyme-linked immunospot (ELISpot) assays showed that, following LK101 administration, neoantigen-specific T-cell responses expanded by a median of 30-fold compared with baseline. Patients who achieved PR had significantly higher ELISpot responses. Meanwhile, expansion of CD8+ effector T cells was also highly consistent with clinical response. These findings provide strong evidence that LK101 effectively activates the body’s antitumor immune response, translating into preliminary clinical benefits.
In terms of safety, LK101 was generally well tolerated. Most treatment-emergent adverse events (TEAEs) were Grade 1–2. The incidence of LK101-related Grade 3 or higher TEAEs was only 8.0%, and no TEAEs led to study discontinuation. Although some patients experienced temporary treatment interruptions due to adverse events, most were able to resume treatment.
Sustained DC vaccine stimulation over a period of up to one year may offer patients more durable and deeper tumor control. These findings suggest that DC vaccines have considerable potential for future use in NSCLC maintenance treatment. We have now initiated a randomized Phase II clinical trial nationwide to further confirm the clinical value of this regimen.
Oncology Frontier: In light of relevant research developments, could you discuss the respective advantages and limitations of personalized neoantigen vaccines and universal/shared-antigen vaccines in the development of therapeutic vaccines for NSCLC?
Professor Jianchun Duan: At this year’s ESMO Congress, therapeutic cancer vaccines were among the most closely watched emerging areas. Currently, vaccine development mainly follows two approaches: personalized vaccines and universal vaccines.
Personalized neoantigen vaccines: These require tumor tissue from each patient for sequencing and neoantigen prediction, after which a customized vaccine is developed specifically for that individual. Their principal advantage is their high degree of specificity. Because the targets are derived from mutations in the patient’s own tumor, they have a strong potential to induce an autologous immune response and precisely activate T-cell immunity.
However, this approach also has limitations. It requires a substantial amount of tumor tissue, while sequencing and testing, manufacturing timelines, and financial costs can all be considerable.
Universal/shared-antigen vaccines: These are generally designed around common tumor antigens identified in public databases and may incorporate dozens or even more potential targets. Their main advantages are convenience and accessibility. They do not require complex individualized testing for every patient and can be manufactured centrally on an industrial scale in large batches, substantially reducing both production time and financial costs.
Their limitation, however, lies in the unpredictability of individual patient coverage. Because of tumor heterogeneity, we cannot guarantee that a particular patient’s tumor will express the antigen targets included in the vaccine. Although increasing the number of targets may expand the potential patient population that could benefit, the degree of antigen matching for each individual still requires further investigation and validation.
