Cancer immunotherapy has been heralded as a breakthrough cancer treatment demonstrating tremendous success in improving tumor responses and survival of patients with hematological cancers and solid tumors. This novel promising treatment approach has in particular triggered optimism for triple negative breast cancer (TNBC) treatment, a subtype of breast cancer with distinct clinical features and poor clinical outcome. In early 2019, the FDA granted the first approval of immune checkpoint therapy, targeting PD-L1 (Atezolizumab) in combination with chemotherapy for the treatment of patients with locally advanced or metastatic PD-L1 positive TNBC. The efficacy of immuno-based interventions varies across cancer types and patient cohorts, which is attributed to a variety of lifestyle, clinical, and pathological factors. For instance, obesity has emerged as a risk factor for a dampened anti-tumor immune response and increased risk of immunotherapy-induced immune-related adverse events (irAEs) but has also been linked to improved outcomes with checkpoint blockade. Given the breadth of the rising global obesity epidemic, it is imperative to gain insight into the immunomodulatory effects of obesity in the peripheral circulation and within the tumor microenvironment. In this review, we resolve the impact of obesity on breast tumorigenesis and progression on the one hand, and on the immune contexture on the other hand. Finally, we speculate on the potential implications of obesity on immunotherapy response in breast cancer. This review clearly highlights the need for in vivo obese cancer models and representative clinical cohorts for evaluation of immunotherapy efficacy.

Background

Groundbreaking studies have linked the gut microbiome with immune homeostasis and antitumor immune responses. Mounting evidence has also demonstrated an intratumoral microbiome, including in soft tissue sarcomas (STS), although detailed characterization of the STS intratumoral microbiome is limited. We sought to characterize the intratumoral microbiome in patients with STS undergoing preoperative radiotherapy and surgery, hypothesizing the presence of a distinct intratumoral microbiome with potentially clinically significant microbial signatures.

Methods

We prospectively obtained tumor and stool samples from adult patients with non-metastatic STS using a strict sterile collection protocol to minimize contamination. Metagenomic classification was used to estimate abundance using genus and species taxonomic levels across all classified organisms, and data were analyzed with respect to clinicopathologic factors.

Results

Fifteen patients were enrolled. Most tumors were located at an extremity (67%) and were histologic grade 3 (87%). 40% were well-differentiated/dedifferentiated liposarcoma histology. With a median follow-up of 24 months, 4 (27%) patients developed metastases, and 3 (20%) died. Despite overwhelming human DNA (>99%) intratumorally, we detected a small but consistent proportion of bacterial DNA (0.02-0.03%) in all tumors, including Proteobacteria, Bacteroidetes, and Firmicutes, as well as viral species. In the tumor microenvironment, we observed a strong positive correlation between viral relative abundance and natural killer (NK) infiltration, and higher NK infiltration was associated with superior metastasis-free and overall survival by immunohistochemical, flow cytometry, and multiplex immunofluorescence analyses.

Conclusions

We prospectively demonstrate the presence of a distinct and measurable intratumoral microbiome in patients with STS at multiple time points. Our data suggest that the STS tumor microbiome has prognostic significance with viral relative abundance associated with NK infiltration and oncologic outcome. Additional studies are warranted to further assess the clinical impact of these findings.