MASCT-I-1005 was a single-center, open-label, 2-part, phase 1 trial (clinicaltrials.gov, NCT04074564). Part A assessed the safety and efficacy of two administration schedules for MASCT-I when combined with camrelizumab and apatinib at fixed doses in patients with advanced bone and soft-tissue sarcomas, and part B further assessed MASCT-I (using the schedule selected based on part A) in combination with apatinib at fixed dose. Here, we report the findings in part A.

Eligible patients were 14–70 years of age; had histologically and cytologically confirmed unresectable recurrent or metastatic bone and soft-tissue sarcoma according to the World Health Organization Classification of Soft Tissue and Bone Tumors; had progressed on at least one line of anti-tumor therapy (such as anthracycline-based chemotherapy or VEGFR-TKI) based on Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (for alveolar soft part sarcoma [ASPS] and clear cell sarcoma patients, no prior treatment was allowed for recruitment as no standard treatment existed; for patients with metastases, the maximum diameter should be no more than 8 cm); had at least one measurable lesion per RECIST version 1.1; had an Eastern Cooperative Oncology Group (ECOG) performance status of 0–1 (for amputees, performance status of 2 was allowed for recruitment); had a life expectancy of at least 6 months; had adequate cardio-pulmonary function; and had adequate hematological, hepatic, and renal function. The key exclusion criteria included active bone metastases or brain metastases; active or history of autoimmune diseases or syndrome, or requiring chronic use of steroids, immunomodulators or immunosuppressive drugs within 2 weeks before study entry; anti-tumor therapies such as chemotherapy, radiation therapy, or targeted therapy within 4 weeks prior to study entry; prior treatment with MASCT, other cellular immunotherapy, or antibodies against PD-1, programmed cell death ligand 1 (PD-L1), or cytotoxic T lymphocyte-associated antigen-4 within the past 1 year; uncontrolled medical disorder (including active tuberculosis, hepatitis B, hepatitis C, human immunodeficiency virus, or syphilis infection); and being pregnant or planning to be pregnant.

MASCT-I cells were manufactured by HRYZ Biotech Co. in a Good Manufacturing Practice grade facility according to the manufacturing protocol as previously described with some modifications [15, 16]. Briefly, peripheral blood mononuclear cells (PBMCs) were collected via leukapheresis from each patient and isolated by density gradient centrifugation using Lymphoprep (Nycomed Pharma) before each course of immunotherapy. Isolated PBMCs were cryopreserved in liquid nitrogen before used for DC and T cell product preparation.

For dendritic cell preparation, PBMCs were thawed and incubated in a culture flask (Corning) at 37 °C, 5% CO2 for 30–60 min. Non-adhesive cells were then removed, and adherent monocytes were cultured in AIM-V medium (Gibco) supplemented with GM-CSF (1000 U/mL) and IL-4 (500 U/mL). On day 5, immature DCs were pulsed with multiple antigen peptides pool (1 μg/mL/peptide) for 24 h. The DCs were then matured with a DC maturation cocktail containing monophosphoryl lipid A, IFNγ, and prostaglandin E2 for 48 h to generate multiple antigen-loaded mature DCs.

To prepare tumor antigen-specific T cells, the frozen PBMCs were thawed and co-cultured with antigen-loaded mature DCs (as described above) in the presence of IL-2 (1000 IU/mL), IL-7 (10 ng/mL), IL-15 (10 ng/mL), IL-21(30 ng/mL), and anti-PD-1 antibody (7.5 μg/mL) for 5 days, followed by T cell expansion with anti-CD3 antibody (50 ng/mL) for another 2 days. The culture medium was then replaced with fresh AIM-V medium supplemented with IL-2, IL-7, IL-15, IL21, and anti-PD-1 antibody, and cultured for another 1–2 weeks to obtain tumor antigen-specific T cells.

The quality of DCs and T cells should meet the specification-releasing of MASCT-I before releasing for cell infusion.

Eligible patients were randomly assigned (1:1) to 2 groups based on different administration schedules of MASCT-I in the first course, by the randomization specialist of Shanghai Canming Medical Technology Co., Ltd, using a central block randomization method. One MASCT-I course consisted of 3 DC subcutaneous injections and 3 active T cell infusions, with each T cell infusion being administered within 18–27 days after each DC injection (Additional file Fig. S1). In schedule-I group, 3 DC injections were administered with a 28-day interval in all courses (i.e., intermittent DC injection and T cell infusion for 3 times in each course); in schedule-II group, 3 DC injections were administered with a 7-day interval in the first course and with a 28-day interval thereafter (i.e., 3 consecutive DC injections followed by 3 consecutive T cell reinfusions in the first course and intermittent DC injection and T cell infusion for 3 times in other courses).

In both groups, 1–2 days after the first apheresis, patients received 1-h intravenous infusion of camrelizumab 200 mg every 3 weeks and oral apatinib 250 mg daily in 28-day cycles.

To manage toxicity, treatment interruption of camrelizumab and treatment interruption and/or dose reduction of apatinib (250 mg every 2 days or 125 mg daily) were allowed (see Additional file 1: Supplementary method for details).

The triple therapy was discontinued upon disease progression or recurrent, intolerable toxicity, withdrawal of consent, pregnancy, substantial noncompliance with study requirements, interruption of study treatment for more than 3 weeks, initiation of new anti-tumor treatment, loss to follow-up, or death. For patients not tolerate camrelizumab or apatinib, MASCT-I could be continued until disease progression or intolerance to MASCT-I. Patients who had radiological disease progression were permitted to continue study treatment if the investigator judged that the patients would benefit from and were tolerant to the continued treatment.

The primary endpoint was safety of MASCT-I in combination with camrelizumab plus apatinib in patients with advanced bone and soft-tissue sarcoma. The secondary endpoints included ORR, disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and immune response.

Safety assessments, including monitoring for adverse events (AEs), were done from the signing of informed consent until the end of study treatment and 4 weeks after the last administration of study treatment. AEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0. The tumor response was assessed by computed tomography or magnetic resonance imaging scan every 8 weeks according to RECIST version 1.1. Survival follow-up was done 4 weeks after the study treatment, every 8 weeks for the first 3 months, and then every 12 weeks until the end of the study or death.

Tumor-associated antigen specific immune response was detected by using ELISPOT (U-CyTech Biosciences, Netherlands). For patients in the schedule-I group, ELISPOT was done during the first apheresis, before the second DC cell reinfusion in the first course, before the first DC cell reinfusion in the subsequent courses, and at the end of the study treatment. For patients in the schedule-II group, ELISPOT was done during the first apheresis, before the first T cell reinfusion in the first course, before the first DC cell reinfusion in the subsequent courses, and at the end of the study treatment. The threshold for a positive response antigen was set to a net increase of 5 spots per 2 × 105 cells from baseline. Immune responders were defined as patients who generated positive immune response against at least one antigen at one ELISPOT assay after the MASCT-I treatment.

Due to the exploratory nature of this phase 1 study, sample sizes of the 2 parts were not determined on the basis of statistical hypotheses. For part A reported here, a total of 20 patients were planned.

Safety analyses were done in patients who received study treatment and had at least one record of safety assessment. Efficacy analyses were done in the intention-to-treat population including all patients enrolled. ORR and DCR were reported, and their 95% CI were calculated via the Clopper-Pearson method. Kaplan–Meier method was used to estimate the median PFS and OS, with 95% CIs obtained by means of Greenwood’s formula.