Soft-tissue sarcomas represent a heterogeneous group of cancer, with more than 50 histological subtypes^1,2. The clinical presentation of patients with different subtypes is often atypical, and responses to therapies such as immune checkpoint blockade vary widely^3,4. To explain this clinical variability, here we study gene expression profiles in 608 tumours across subtypes of soft-tissue sarcoma. We establish an immune-based classification on the basis of the composition of the tumour microenvironment and identify five distinct phenotypes: immune-low (A and B), immune-high (D and E), and highly vascularized (C) groups. In situ analysis of an independent validation cohort shows that class E was characterized by the presence of tertiary lymphoid structures that contain T cells and follicular dendritic cells and are particularly rich in B cells. B cells are the strongest prognostic factor even in the context of high or low CD8⁺ T cells and cytotoxic contents. The class-E group demonstrated improved survival and a high response rate to PD1 blockade with pembrolizumab in a phase 2 clinical trial. Together, this work confirms the immune subtypes in patients with soft-tissue sarcoma, and unravels the potential of B-cell-rich tertiary lymphoid structures to guide clinical decision-making and treatments, which could have broader applications in other diseases.

Immune checkpoint inhibitor (ICI) monoclonal antibodies (mAbs) targeting cytotoxic T-lymphocyte antigen 4 (CTLA-4), programmed cell death protein 1 (PD-1) or its ligand (PD-L1) produce unique toxicity profiles. The objective of this review was to identify patterns and incidence of immune-related adverse events (irAE) based on tumour type and ICI class. Medline, EMBASE and COCHRANE databases were searched to identify prospective monotherapy trials of ICIs from 2003 to November 2015. Paired reviewers selected studies for inclusion and extracted data. Odds ratio (OR), χ² tests and multivariable regression models were used to analyse for effect size and associations. We identified 48 trials (6938 patients), including 26 CTLA-4, 17 PD-1, 2 PD-L1 trials, and 3 studies tested both CTLA-4 and PD-1. Grade 3/4 irAE were more common with CTLA-4 mAbs compared with PD-1 (31% versus 10%). All grades colitis (OR 8.7, 95% CI 5.8–12.9), hypophysitis (OR 6.5, 95% CI 3.0–14.3) and rash (OR 2.0, 95% CI 1.8–2.3) were more frequent with CTLA-4 mAbs; whereas pneumonitis (OR 6.4, 95% CI 3.2–12.7), hypothyroidism (OR 4.3, 95% CI 2.9–6.3), arthralgia (OR 3.5, 95% CI 2.6–4.8) and vitiligo (OR 3.5, 95% CI 2.3–5.3) were more common with PD-1 mAbs. Comparison of irAE from the three most studied tumour types in PD-1 mAbs trials [melanoma (n = 2048), non-small-cell lung cancer (n = 1030) and renal cell carcinoma (n = 573)] showed melanoma patients had a higher frequency of gastrointestinal and skin irAE and lower frequency of pneumonitis. CTLA-4 and PD-1 mAbs have distinct irAE profiles. Different immune microenvironments may drive histology-specific irAE patterns. Other tumour-dependent irAE profiles may be identified as data emerge from ICI trials.

There are limited data about the quality of immune-related adverse event (irAE) reporting in immune checkpoint inhibitor (ICI) clinical trial publications. A systematic search of citations from Medline, EMBASE and Cochrane databases identified prospective clinical trials involving ICIs in advanced solid tumors from 2003 to 2013. A 21-point quality score (QS) was adapted from the CONSORT harms extension statement. Linear regression was used to identify factors associated with quality reporting. After a review of 2628 articles, 50 trial reports were included, with ICIs as either monotherapy (54%) or part of a combination regimen (46%). The mean QS was 11.21 points (range 3.50–17.50 points). The median grade 3/4 AE rate reported was 21% (range 0%–66%) and 29/50 (58%) trials concluded that irAEs were tolerable. Multivariate regression analysis revealed that year of publication (within last 5 years, P = 0.01) and journal impact factor >15 (P = 0.004) were associated with higher QS. Complete reporting of specific characteristics of irAEs including onset, management and reversibility were reported by 14%, 8% and 6% of studies, respectively. The incidence of grade 3/4 adverse events was higher for inhibitors against CTLA-4 compared with other immune checkpoints (P < 0.001). The reporting of irAEs is suboptimal. A standardized reporting method of irAEs that accounts for tolerability, management and reversibility is needed and would enable a more precise evaluation of the therapeutic risk benefit ratio of ICIs.

Whereas T cells have been considered the major immune cells of the tumor microenvironment able to induce tumor regression and control cancer clinical outcome, a burst of recent publications pointed to the fact that B cells may also play a prominent role. Activated in germinal centers of tertiary lymphoid structures, B cells can directly present tumor-associated antigens to T cells or produce antibodies that increase antigen presentation to T cells or kill tumor cells, resulting in a beneficial clinical impact. Immune complexes can also increase inflammation, angiogenesis, and immunosuppression via macrophage and complement activation, resulting in deleterious impact.

The majority of published studies in recurrent or metastatic nasopharyngeal carcinoma (RM-NPC) are single-arm trials. Reliable modelling of progression-free survival (PFS) and overall survival (OS) outcomes, therefore, is difficult. This study aim to analyse existent literature to estimate the relative efficacy of available systemic regimens in RM-NPC, as well as provide estimates of aggregate OS and PFS. We conducted a systematic search of MEDLINE, EMBASE and the Cochrane Library to March 2015. Clinical trials (in English only) investigating cytotoxic and molecularly targeted agents in adult patients with RM-NPC were included. All relevant studies were assessed for quality using Downs and Blacks (DB) checklist (maximum quality score of 27). Aggregate data analysis and Student’s t-test were performed for all identified studies (model A). For studies that published analysable Kaplan−Meier curves, survival data were extracted and marginal proportional hazards models were constructed (model B). A total of 56 studies were identified and included in model A, 26 of which had analysable Kaplan−Meier curves and were included in model B. The 26 studies in model B had significantly higher mean DB scores than the remaining 30 (17.3 vs 13.7, P=0.002). For patients receiving first line chemotherapy, the estimated median OS was 15.7 months by model A (95% CI, 12.3–19.1), and 19.3 months by model B (95% CI, 17.6–21.1). For patients undergoing second line or higher therapies (2nd+), the estimated median OS was 11.5 months by model A (95% CI 10.1–12.9), and 12.5 months by model B (95% CI 11.9–13.4). PFS estimates for patients undergoing first-line chemotherapy by model A was 7.6 months (95% CI, 6.2–9.0), and 8.0 months by model B (95% CI, 7.6–8.8). For patients undergoing therapy in the 2nd+ setting, the estimated PFS by model A was 5.4 months (95% CI, 3.8–7.0), and 5.2 months by model B (95% CI, 4.7–5.6). We present the first aggregate estimates of OS and PFS for RM-NPC patients receiving first and second-line or higher treatment settings, which could inform the design of future clinical trials in this disease setting.