Most lower-extremity assistive robots are designed to actively assist gait1-7 without considering long-term neuromuscular adaptations8-11. Here we present a lightweight (0.96 kg) robot that administers isokinetic resistance training to sustain neuromuscular rehabilitation after removal. The device integrates a variable stiffness mechanism with a back-drivable damping motor to make available safe, portable and customizable resistance training to juveniles with spinal muscular atrophy (SMA) type II. In a study involving six such juvenile participants, substantial improvements in lower-extremity motor ability were observed after 6 weeks of robot-assisted training in a clinical trial (NCT06648486). Participants gained the ability to perform sit-to-stand transitions with hands on knees but without external support from an average seated knee flexion angle of 111° to 104°, representing a 7° improvement from pre-intervention. This improvement was accompanied by greatly increased bilateral knee joint function (peak torque: +130%; range of motion (ROM): +51%; work: +97%). Marked physiological quadriceps muscle hypertrophy was observed (anatomical cross-sectional area (ACSA): +12%; volume: +19%; physiological cross-sectional area (PCSA): +21%) alongside enhanced femoral nerve conduction (compound muscle action potential (CMAP): +19%), representing physiological changes consistent with the observed functional improvements. Notably, participants were able to retain their gains after discontinuing isokinetic training and returning to their conventional physiotherapy routines. These results indicate that even temporary exposure to isokinetic resistance training through wearable robotics may facilitate enduring neuromuscular recovery.

Intravenous administration of anti-CTLA4 with anti-PD1 provides durable tumour responses but causes severe treatment-related adverse events in patients with cancer1. Intratumoural administration at lower doses but high local concentrations could enhance antitumour efficacy while minimizing systemic exposure and toxicity. Here we report the randomized multicentre phase 1b NIVIPIT trial (ClinicalTrials.gov: NCT02857569 ), which enrolled 61 patients with untreated metastatic melanoma, randomly assigned 2:1 to receive intravenous nivolumab (anti-PD1; 1 mg kg-1) combined with either intratumoural ipilimumab (anti-CTLA4; 0.3 mg kg-1) or intravenous ipilimumab (3 mg kg-1). The primary end-point was met with significantly lower incidence of grade 3 or 4 treatment-related adverse events at 6 months in the intratumoural versus intravenous arm (22.6% versus 57.1%), equivalent to anti-PD1 monotherapy. RECIST (response evaluation criteria in solid tumours) best objective response rate reached 65.7% for anti-CTLA4 injected lesions and 50% for uninjected lesions, confirming the relationship between intratumoural exposure to anti-CTLA4 and efficacy. Baseline tumour immune profiling revealed that protumoural activated regulatory T (Treg) cells and M2 macrophages predict durable clinical benefit, regardless of the anti-CTLA4 administration route. A decrease in activated intratumoural Treg cells occurred only in patients who showed durable clinical benefit, who also presented high intratumoural Fcγ receptor (FcγR) expression. Our results provide a rationale for intratumoural anti-CTLA4 strategies in oligometastatic and early-stage cancers and indicate that high intratumoural activated Treg cell and FcγR+ M2 macrophage numbers are prerequisites for efficacy of combined anti-CTLA4 and anti-PD1.

Netrin1, a developmental cue, is a master regulator of tumour epithelial-to-mesenchymal transition (EMT)1, a mechanism that is known to drive resistance to chemotherapy2. A netrin1 antibody (NP137)3 has been shown to inhibit tumour EMT in preclinical1 and clinical4 settings. In animal models of pancreatic cancer, netrin1 and its receptor neogenin have been shown to promote tumour progression5, EMT5 and metastasis6. Here we report the results of a phase 1b study that assesses the combination of NP137 with modified FOLFIRINOX (mFOLFIRINOX) in first line patients with locally advanced pancreatic cancer (ClinicalTrials.gov: NCT05546853 ). Forty-three patients were enrolled and received mFOLFIRINOX plus NP137 every other week for up to 12 cycles. NP137 was well tolerated. Median progression-free survival (PFS) was 10.85 months (95% confidence interval, 10.03-15.61) and median overall survival was 16.43 months (95% confidence interval, 12.75-non-reached), with 21 patients remaining alive at the time of data cut-off. Post-therapy conversion surgery occurred in 23% of patients. Laser capture microdissection was performed on pre-therapeutic biopsies and surgical specimens. Microbulk RNA sequencing confirmed that the main pathway that was down-regulated with the combination of mFOLFIRINOX plus NP137 was EMT. Moreover, survival outcomes were extended for patients with tumour cells that expressed high levels of the netrin1 receptor neogenin-median PFS 15.65 months in neogenin-high versus 10.22 months in neogenin low. Our results support the idea that netrin1 blockade alleviates resistance to chemotherapy by inhibiting EMT, particularly in neogenin-high pancreatic cancer.

Autosomal recessive deafness 9, caused by OTOF gene mutations, is characterized by severe-to-complete congenital deafness1. Although gene therapy has shown benefits in a small number of patients2-5, its safety and efficacy across broader age ranges and longer follow-up periods, as well as predictors of treatment outcomes, remain unclear. In this single-arm, multicentre trial conducted at eight centres, 42 participants (aged 0.8-32.3 years) received adeno-associated virus (AAV) serotype 1 carrying a human OTOF coding transgene (AAV1-hOTOF) at three vector dose groups, with up to 2.5-year follow-up. The primary end point was dose-limiting toxicity within 6 weeks. The secondary end point assessed efficacy and adverse events. No dose-limiting toxicities were observed. Grade 3 adverse events included decreased neutrophil count. Hearing was recovered in 90% of participants treated with AAV1-hOTOF, with gradual and stable improvement in auditory brainstem response threshold from greater than 97 ± 1 dB normalized hearing level at baseline to 54 ± 3, 51 ± 3, 50 ± 3 and 42 ± 5 dB normalized hearing level at 1, 1.5, 2 and 2.5 years, respectively, and behavioural audiometry improving from greater than 96 ± 3 dB hearing level at baseline to 37 ± 5 dB hearing level at 2.5 years. Participants aged 0.5-18 years showed greater hearing improvement than adults. A higher number of present distortion product otoacoustic emissions at baseline or biallelic non-truncated OTOF variants was associated with better hearing recovery. Participants with hearing recovery demonstrated gradual improvement in speech perception. AAV1-hOTOF is well-tolerated and efficacious across a broader patient population, with sustained therapeutic benefits for up to 2.5 years. Chinese Clinical Trial Registry registration: ChiCTR2200063181 .

β-Thalassaemia is caused by reduced or absent production of β-haemoglobin1-4. Previously, we performed laboratory-scale electroporation of CD34+ haematopoietic stem and progenitor cells from patients with β-thalassaemia using a transformer base editor5,6. The aim was to target the binding motif of the transcription repressor BCL11A in the HBG1 and HBG2 promoters7 to reactivate fetal haemoglobin (HbF) production. Here we present results of a phase 1 clinical trial (ClinicalTrials.gov identifier: NCT06024876) of five patients who received autologous CD34+ cells modified using a transformer base editor at clinical scale (CS-101). With a median follow-up of 23.0 months after CS-101 infusion, the median times to neutrophil and platelet engraftment were 16 days and 25 days, respectively. Moreover, all patients had stopped red blood cell transfusions, with a median time to the last transfusion of 18 days after CS-101 infusion. The mean total haemoglobin and HbF concentrations were 12.4 ± 1.0 and 11.5 ± 0.9 g dl-1, respectively, at month 3 after infusion. These levels remained at similar or higher levels throughout the follow-up period, which indicated rapid haematopoietic reconstitution. The adverse events of CS-101 were generally consistent with those of busulfan myeloablative conditioning and autologous haematopoietic stem and progenitor cell transplantation. No deaths or cancer occurrences were reported. In summary, CS-101 can lead to rapid and sustained increases in both total haemoglobin and HbF levels, which resulted in early and enduring transfusion independence.

Despite promising data showing that circulating tumour DNA (ctDNA) dynamics during treatment can inform real-time tumour response and recurrence risk1, how best to translate these insights into actionable clinical decision-making remains unclear. Here we report results from the EP-STAR trial-a multi-centre, ctDNA-driven, risk-adapted, non-randomized phase II study ( NCT04072107 ; ClinicalTrials.gov) testing whether a risk-adaptive treatment (RAT) strategy guided by on-treatment ctDNA dynamics can meaningfully improve survival, using nasopharyngeal carcinoma as a model. Eligible patients were enrolled and began treatment with standard-of-care gemcitabine-cisplatin neoadjuvant chemotherapy (GP-NAC; the P in this abbreviation stands for platinum)2, followed by RAT or standard-of-care chemoradiotherapy guided by ctDNA clearance trajectory during GP-NAC. Protocol-eligible patients who did not receive RAT, drawn from a prospectively registered ctDNA biomarker cohort ( NCT03855020 )3, served as a non-randomized, contemporaneous no-RAT external cohort. The primary end-point was failure-free survival (FFS) in the RAT group. After a median follow-up of 47.3 months, the 3-year FFS was 89.1% (83.2-95.0%) in the RAT group (n = 110). Patients who received RAT showed significantly improved FFS (P = 0.003, log-rank test) compared with the no-RAT external cohort (hazard ratio = 0.41 [0.23-0.75]; P = 0.004, Cox regression model). The RAT strategy was well-tolerated with no treatment-related deaths. Collectively, these data show that a ctDNA-driven RAT paradigm could be a promising strategy to improve survival, challenging the conventional fixed-course, static treatment approach.

Globally, as many as 12 million girls marry before the age of 18 every year; in northern Nigeria, 80% of girls marry before 18 (refs. 1,2). Although such marriages may be deemed the best available option by many girls and parents, numerous studies suggest that, when delayed marriage is made possible, it benefits educational attainment, improves health by reducing maternal mortality and morbidity, and leads to many other benefits to girls' lives3-8. Despite this, little is known about what reduces child marriage, and successful interventions tend to have an impact of just a few percentage points. We use a paired cluster-randomized trial in 18 communities to rigorously evaluate a locally tailored big-push intervention called Pathways to Choice in northern Nigeria. We show that Pathways decreases rates of marriage among adolescent girls from 86% in the control group to only 21% in the treatment group-just over an 80% decrease. Although a key part of Pathways' effect is a significant increase in girls re-enrolling in school, education alone cannot explain its effects on child marriage. We argue that Pathways' whole-community focus reduces the likelihood of social backlash and contributes meaningfully to its success. Our results demonstrate that a big push can significantly alter entrenched, normative behaviour around child marriage, and that bundled interventions may be greater than the sum of their parts.

Mazdutide is a once-weekly glucagon and glucagon-like peptide 1 receptor dual agonist developed for the treatment of type 2 diabetes (T2D)1. Here we report on a randomized phase III trial assessing the efficacy and safety of mazdutide, compared with dulaglutide, in participants with T2D who were also treated with background oral anti-diabetic drugs. In this study, 731 participants with T2D were randomized 1:1:1 to receive 4 mg mazdutide, 6 mg mazdutide or 1.5 mg dulaglutide for 28 weeks. Both doses of mazdutide showed non-inferiority and superiority to the 1.5-mg dose of dulaglutide in terms of the mean change in the diagnostic marker glycated haemoglobin A1c (HbA1c) from baseline to week 28, with a least-squares mean treatment difference of -0.24% (P = 0.0032) for 4 mg mazdutide and -0.30% (P = 0.0003) for 6 mg mazdutide, relative to 1.5 mg dulaglutide. Significantly greater reductions in body weight were achieved with mazdutide than with dulaglutide, with a least-squares mean treatment difference of -3.78% for 4 mg mazdutide and -5.76% for 6 mg mazdutide (both P < 0.0001), relative to dulaglutide. Moreover, significantly more participants who received mazdutide 4 mg or 6 mg reached the composite end point of HbA1c < 7.0% with a body-weight reduction of at least 5% at week 28 (both P < 0.0001), compared with those who received dulaglutide. The most common treatment-emergent adverse events were diarrhoea, nausea and vomiting. In summary, we found that in Chinese participants with T2D, 28 weeks of treatment with mazdutide (4 mg and 6 mg) provided reductions in HbA1c and body weight that were superior to those attained with 1.5 mg dulaglutide. Mazdutide was generally safe, although the incidence of gastrointestinal adverse events was higher for mazdutide than for dulaglutide.

Despite advances in type 2 diabetes (T2D) management, unmet needs remain for therapies that effectively control hyperglycaemia while addressing comorbid metabolic disorders1,2. Here we assessed the efficacy and safety of the dual glucagon receptor (GCGR)/glucagon-like peptide-1 receptor (GLP-1R) agonist mazdutide monotherapy versus placebo in Chinese adults with T2D controlled inadequately with diet and exercise alone. In this phase 3 trial, 320 participants (mean glycated haemoglobin A1c (HbA1c) of 8.24%, body mass index of 28.2 kg m-2 and diabetes duration of 1.9 years) were randomized 1:1:1 to receive weekly subcutaneous injections of mazdutide (4 mg or 6 mg) or placebo for 24 weeks, followed by a 24-week extended mazdutide treatment. At week 24, mazdutide significantly reduced HbA1c versus placebo (primary endpoint): -1.57% with mazdutide 4 mg and -2.15% with mazdutide 6 mg, versus -0.14% with placebo, with treatment differences of -1.43% and -2.02% (both P < 0.0001). Weight loss from baseline at week 24 occurred with -5.61% (4 mg) and -7.81% (6 mg) versus -1.26% (placebo) (both P < 0.0001). Furthermore, more participants with mazdutide achieved HbA1c < 7.0%, weight loss ≥ 5% (all P < 0.0001) and composite endpoints (HbA1c < 7.0% and weight loss ≥ 5%) versus placebo (P = 0.0006 for 4 mg; P < 0.0001 for 6 mg) at week 24. The most common adverse events-diarrhoea, decreased appetite and nausea-were consistent with GLP-1R agonists. These results establish mazdutide monotherapy as an effective intervention providing clinically meaningful glycaemic control and weight reduction alongside a favourable safety profile in this population.

The identification of therapeutic strategies to induce sustained antiretroviral therapy (ART)-free control of HIV infection is a major priority1. Combination immunotherapy including HIV vaccination, immune stimulation, latency reversal and passive transfer of broadly neutralizing antibodies (bNAbs) has shown promise in non-human primate models2-6, but few studies have translated such approaches into people. Here we performed a single-arm, proof-of-concept study in ten people living with HIV on ART, combining the following three approaches: (1) therapeutic vaccination with an HIV Gag conserved element-targeted DNA + IL-12 prime/modified vaccinia Ankara (MVA) boost regimen followed by (2) administration of two bNAbs (10-1074, VRC07-523LS) and a toll-like receptor 9 agonist (lefitolimod) during ART suppression, followed by (3) repeat bNAb administration at the time of ART interruption (Clinicaltrials.gov: NCT04357821 ). Seven out of the ten participants exhibited post-intervention control after pausing ART, independent of residual bNAb plasma levels. Robust expansion of activated CD8+ T cells early in response to rebounding virus correlated with a lower median viral load after peak viraemia off ART. These data suggest that combination immunotherapy approaches might prove effective in inducing sustained control of HIV by slowing rebound and improving CD8+ T cell responses, and that these approaches should continue to be optimized.

Immune checkpoint blockade has led to paradigm shifts in the treatment of various tumour types1-4, yet limited efficacy has been observed in patients with metastatic mismatch-repair-proficient (pMMR) colorectal cancer5. Here we report clinical results and in-depth analysis of patients with early-stage pMMR colon cancer from the phase II NICHE study (ClinicalTrials.gov: NCT03026140). A total of 31 patients received neoadjuvant treatment of nivolumab plus ipilimumab followed by surgery. The response rate was 26% and included six patients with a major pathological response (10% or less residual viable tumour). One patient with an ongoing clinical complete response did not undergo surgery. Circulating tumour DNA was positive in 26 of 31 patients at baseline, and clearance was observed in 5 of 6 responders before surgery, whereas 19 of 20 non-responders remained circulating tumour DNA positive. Responses were observed despite a low tumour mutational burden in all tumours, whereas chromosomal genomic instability scores were significantly higher in responders than in non-responders. Furthermore, responding tumours had significantly higher baseline expression of proliferation signatures and TCF1, and imaging mass cytometry revealed a higher percentage of Ki-67+ cancer and Ki-67+CD8+ T cells in responders than in non-responders. These results provide a comprehensive analysis of response to neoadjuvant immune checkpoint blockade in early-stage pMMR colon cancers and identify potential biomarkers for patient selection.

Decades of research have demonstrated that recovery from serious neurological injury will require synergistic therapeutic approaches. Rewiring spared neural circuits after injury is a long-standing goal of neurorehabilitation1,2. We hypothesized that combining intensive, progressive, task-focused training with real-time closed-loop vagus nerve stimulation (CLV) to enhance synaptic plasticity3 could increase strength, expand range of motion and improve hand function in people with chronic, incomplete cervical spinal cord injury. Here we report the results from a prospective, double-blinded, sham-controlled, randomized study combining gamified physical therapy using force and motion sensors to deliver sham or active CLV (ClinicalTrials.gov identifier NCT04288245). After 12 weeks of therapy composed of a miniaturized implant selectively activating the vagus nerve on successful movements, 19 people exhibited a significant beneficial effect on arm and hand strength and the ability to perform activities of daily living. CLV represents a promising therapeutic avenue for people with chronic, incomplete cervical spinal cord injury.

Parkinson's disease is a progressive neurodegenerative condition with a considerable health and economic burden1. It is characterized by the loss of midbrain dopaminergic neurons and a diminished response to symptomatic medical or surgical therapy as the disease progresses2. Cell therapy aims to replenish lost dopaminergic neurons and their striatal projections by intrastriatal grafting. Here, we report the results of an open-label phase I clinical trial (NCT04802733) of an investigational cryopreserved, off-the-shelf dopaminergic neuron progenitor cell product (bemdaneprocel) derived from human embryonic stem (hES) cells and grafted bilaterally into the putamen of patients with Parkinson's disease. Twelve patients were enrolled sequentially in two cohorts-a low-dose (0.9 million cells, n = 5) and a high-dose (2.7 million cells, n = 7) cohort-and all of the participants received one year of immunosuppression. The trial achieved its primary objectives of safety and tolerability one year after transplantation, with no adverse events related to the cell product. At 18 months after grafting, putaminal 18Fluoro-DOPA positron emission tomography uptake increased, indicating graft survival. Secondary and exploratory clinical outcomes showed improvement or stability, including improvement in the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) Part III OFF scores by an average of 23 points in the high-dose cohort. There were no graft-induced dyskinesias. These data demonstrate safety and support future definitive clinical studies.

Parkinson's disease is caused by the loss of dopamine neurons, causing motor symptoms. Initial cell therapies using fetal tissues showed promise but had complications and ethical concerns1-5. Pluripotent stem (PS) cells emerged as a promising alternative for developing safe and effective treatments6. In this phase I/II trial at Kyoto University Hospital, seven patients (ages 50-69) received bilateral transplantation of dopaminergic progenitors derived from induced PS (iPS) cells. Primary outcomes focused on safety and adverse events, while secondary outcomes assessed motor symptom changes and dopamine production for 24 months. There were no serious adverse events, with 73 mild to moderate events. Patients' anti-parkinsonian medication doses were maintained unless therapeutic adjustments were required, resulting in increased dyskinesia. Magnetic resonance imaging showed no graft overgrowth. Among six patients subjected to efficacy evaluation, four showed improvements in the Movement Disorder Society Unified Parkinson's Disease Rating Scale part III OFF score, and five showed improvements in the ON scores. The average changes of all six patients were 9.5 (20.4%) and 4.3 points (35.7%) for the OFF and ON scores, respectively. Hoehn-Yahr stages improved in four patients. Fluorine-18-L-dihydroxyphenylalanine (18F-DOPA) influx rate constant (Ki) values in the putamen increased by 44.7%, with higher increases in the high-dose group. Other measures showed minimal changes. This trial (jRCT2090220384) demonstrated that allogeneic iPS-cell-derived dopaminergic progenitors survived, produced dopamine and did not form tumours, therefore suggesting safety and potential clinical benefits for Parkinson's disease.

At the heart of medicine lies physician-patient dialogue, where skillful history-taking enables effective diagnosis, management and enduring trust1,2. Artificial intelligence (AI) systems capable of diagnostic dialogue could increase accessibility and quality of care. However, approximating clinicians' expertise is an outstanding challenge. Here we introduce AMIE (Articulate Medical Intelligence Explorer), a large language model (LLM)-based AI system optimized for diagnostic dialogue. AMIE uses a self-play-based3 simulated environment with automated feedback for scaling learning across disease conditions, specialties and contexts. We designed a framework for evaluating clinically meaningful axes of performance, including history-taking, diagnostic accuracy, management, communication skills and empathy. We compared AMIE's performance to that of primary care physicians in a randomized, double-blind crossover study of text-based consultations with validated patient-actors similar to objective structured clinical examination4,5. The study included 159 case scenarios from providers in Canada, the United Kingdom and India, 20 primary care physicians compared to AMIE, and evaluations by specialist physicians and patient-actors. AMIE demonstrated greater diagnostic accuracy and superior performance on 30 out of 32 axes according to the specialist physicians and 25 out of 26 axes according to the patient-actors. Our research has several limitations and should be interpreted with caution. Clinicians used synchronous text chat, which permits large-scale LLM-patient interactions, but this is unfamiliar in clinical practice. While further research is required before AMIE could be translated to real-world settings, the results represent a milestone towards conversational diagnostic AI.

Hepatocellular carcinoma remains a life-threatening malignancy with limited therapeutic options following the failure of second-line treatments1,2. Oncolytic viruses selectively replicate in and lyse cancer cells, releasing neoantigens and stimulating systemic antitumour immunity3, offering a potential therapeutic option. Here we present the results of a multicentre phase 1 clinical trial evaluating VG161, an engineered oncolytic herpes simplex virus that expresses IL-12, IL-15, IL-15Rα and a PD-1-PD-L1-blocking fusion protein4, for safety and efficacy in patients with advanced liver cancer. VG161 was well tolerated, with no dose-limiting toxicities observed, and it demonstrated promising efficacy by reshaping the tumour immune microenvironment and re-sensitizing tumours that were previously resistant to systemic treatments. Notably, we also found that patients who had previously been sensitive to checkpoint inhibitor therapy showed enhanced efficacy with VG161 treatment. Furthermore, we developed an efficacy-prediction model based on differentially expressed genes, which successfully identified patients who were likely to benefit from VG161 and predicted prolonged overall survival. These findings position VG161 as a promising third-line therapeutic option for refractory hepatocellular carcinoma. This provides a new avenue for treatment and advances the field of oncolytic virus-based immunotherapies. ClinicalTrials.gov registration: NCT04806464 .

A fundamental challenge for cancer vaccines is to generate long-lived functional T cells that are specific for tumour antigens. Here we find that mRNA-lipoplex vaccines against somatic mutation-derived neoantigens may solve this challenge in pancreatic ductal adenocarcinoma (PDAC), a lethal cancer with few mutations. At an extended 3.2-year median follow-up from a phase 1 trial of surgery, atezolizumab (PD-L1 inhibitory antibody), autogene cevumeran1 (individualized neoantigen vaccine with backbone-optimized uridine mRNA-lipoplex nanoparticles) and modified (m) FOLFIRINOX (chemotherapy) in patients with PDAC, we find that responders with vaccine-induced T cells (n = 8) have prolonged recurrence-free survival (RFS; median not reached) compared with non-responders without vaccine-induced T cells (n = 8; median RFS 13.4 months; P  =  0.007). In responders, autogene cevumeran induces CD8+ T cell clones with an average estimated lifespan of 7.7 years (range 1.5 to roughly 100 years), with approximately 20% of clones having latent multi-decade lifespans that may outlive hosts. Eighty-six percent of clones per patient persist at substantial frequencies approximately 3 years post-vaccination, including clones with high avidity to PDAC neoepitopes. Using PhenoTrack, a novel computational strategy to trace single T cell phenotypes, we uncover that vaccine-induced clones are undetectable in pre-vaccination tissues, and assume a cytotoxic, tissue-resident memory-like T cell state up to three years post-vaccination with preserved neoantigen-specific effector function. Two responders recurred and evidenced fewer vaccine-induced T cells. Furthermore, recurrent PDACs were pruned of vaccine-targeted cancer clones. Thus, in PDAC, autogene cevumeran induces de novo CD8+ T cells with multiyear longevity, substantial magnitude and durable effector functions that may delay PDAC recurrence. Adjuvant mRNA-lipoplex neoantigen vaccines may thus solve a pivotal obstacle for cancer vaccination.

Personalized cancer vaccines (PCVs) can generate circulating immune responses against predicted neoantigens1-6. However, whether such responses can target cancer driver mutations, lead to immune recognition of a patient's tumour and result in clinical activity are largely unknown. These questions are of particular interest for patients who have tumours with a low mutational burden. Here we conducted a phase I trial (ClinicalTrials.gov identifier NCT02950766) to test a neoantigen-targeting PCV in patients with high-risk, fully resected clear cell renal cell carcinoma (RCC; stage III or IV) with or without ipilimumab administered adjacent to the vaccine. At a median follow-up of 40.2 months after surgery, none of the 9 participants enrolled in the study had a recurrence of RCC. No dose-limiting toxicities were observed. All patients generated T cell immune responses against the PCV antigens, including to RCC driver mutations in VHL, PBRM1, BAP1, KDM5C and PIK3CA. Following vaccination, there was a durable expansion of peripheral T cell clones. Moreover, T cell reactivity against autologous tumours was detected in seven out of nine patients. Our results demonstrate that neoantigen-targeting PCVs in high-risk RCC are highly immunogenic, capable of targeting key driver mutations and can induce antitumour immunity. These observations, in conjunction with the absence of recurrence in all nine vaccinated patients, highlights the promise of PCVs as effective adjuvant therapy in RCC.

Cancer immunotherapies with antibodies blocking immune checkpoint molecules are clinically active across multiple cancer entities and have markedly improved cancer treatment1. Yet, response rates are still limited, and tumour progression commonly occurs2. Soluble and cell-bound factors in the tumour microenvironment negatively affect cancer immunity. Recently, growth differentiation factor 15 (GDF-15), a cytokine that is abundantly produced by many cancer types, was shown to interfere with antitumour immune response. In preclinical cancer models, GDF-15 blockade synergistically enhanced the efficacy of anti-PD-1-mediated checkpoint inhibition3. In a first-in-human phase 1-2a study (GDFATHER-1/2a trial, NCT04725474 ), patients with advanced cancers refractory to anti-PD-1 or anti-PD-L1 therapy (termed generally as anti-PD-1/PD-L1 refractoriness) were treated with the neutralizing anti-GDF-15 antibody visugromab (CTL-002) in combination with the anti-PD-1 antibody nivolumab. Here we show that durable and deep responses were achieved in some patients with non-squamous non-small cell lung cancer and urothelial cancer, two cancer entities identified as frequently immunosuppressed by GDF-15 in an in silico screening of approximately 10,000 tumour samples in The Cancer Genome Atlas database. Increased levels of tumour infiltration, proliferation, interferon-γ-related signalling and granzyme B expression by cytotoxic T cells were observed in response to treatment. Neutralizing GDF-15 holds promise in overcoming resistance to immune checkpoint inhibition in cancer.

Interleukin-15 (IL-15) promotes the survival of T lymphocytes and enhances the antitumour properties of chimeric antigen receptor (CAR) T cells in preclinical models of solid neoplasms in which CAR T cells have limited efficacy1-4. Glypican-3 (GPC3) is expressed in a group of solid cancers5-10, and here we report the evaluation in humans of the effects of IL-15 co-expression on GPC3-expressing CAR T cells (hereafter GPC3 CAR T cells). Cohort 1 patients ( NCT02905188 and NCT02932956 ) received GPC3 CAR T cells, which were safe but produced no objective antitumour responses and reached peak expansion at 2 weeks. Cohort 2 patients ( NCT05103631 and NCT04377932 ) received GPC3 CAR T cells that co-expressed IL-15 (15.CAR), which mediated significantly increased cell expansion and induced a disease control rate of 66% and antitumour response rate of 33%. Infusion of 15.CAR T cells was associated with increased incidence of cytokine release syndrome, which was controlled with IL-1/IL-6 blockade or rapidly ameliorated by activation of the inducible caspase 9 safety switch. Compared with non-responders, tumour-infiltrating 15.CAR T cells from responders showed repression of SWI/SNF epigenetic regulators and upregulation of FOS and JUN family members, as well as of genes related to type I interferon signalling. Collectively, these results demonstrate that IL-15 increases the expansion, intratumoural survival and antitumour activity of GPC3 CAR T cells in patients.