Sarcoma microenvironment cell states and ecosystems are associated with prognosis and predict response to immunotherapy.

Characterization of the diverse malignant and stromal cell states that make up soft tissue sarcomas and their correlation with patient outcomes has proven difficult using fixed clinical specimens. Here, we employed EcoTyper, a machine-learning framework, to identify the fundamental cell states and cellular ecosystems that make up sarcomas on a large scale using bulk transcriptomes with clinical annotations. We identified and validated 23 sarcoma-specific, transcriptionally defined cell states, many of which were highly prognostic of patient outcomes across independent datasets. We discovered three conserved cellular communities or ecotypes associated with underlying genomic alterations and distinct clinical outcomes. We show that one ecotype defined by tumor-associated macrophages and epithelial-like malignant cells predicts response to immune-checkpoint inhibition but not chemotherapy and validate our findings in an independent cohort. Our results may enable identification of patients with soft tissue sarcomas who could benefit from immunotherapy and help develop new therapeutic strategies.

Integrated germline and somatic features reveal divergent immune pathways driving ICB response.

Immune Checkpoint Blockade (ICB) has revolutionized cancer treatment, however mechanisms determining patient response remain poorly understood. Here we used machine learning to predict ICB response from germline and somatic biomarkers and interpreted the learned model to uncover putative mechanisms driving superior outcomes. Patients with higher T follicular helper infiltrates were robust to defects in the class-I Major Histocompatibility Complex (MHC-I). Further investigation uncovered different ICB responses in MHC-I versus MHC-II neoantigen reliant tumors across patients. Despite similar response rates, MHC-II reliant responses were associated with significantly longer durable clinical benefit (Discovery: Median OS=63.6 vs. 34.5 months P=0.0074; Validation: Median OS=37.5 vs. 33.1 months, P=0.040). Characteristics of the tumor immune microenvironment reflected MHC neoantigen reliance, and analysis of immune checkpoints revealed LAG3 as a potential target in MHC-II but not MHC-I reliant responses. This study highlights the value of interpretable machine learning models in elucidating the biological basis of therapy responses.

Application of a Machine Learning Algorithm in Prediction of Abusive Head Trauma in Children.

PURPOSE: We explored the application of a machine learning algorithm for the timely detection of potential abusive head trauma (AHT) using the first free-text note of an encounter and demographic information. METHODS: First free-text physician notes and demographic information were collected for children under 5 years of age at a Level 1 Trauma Center. The control group, which included patients with head/neck injury, was compared to those with AHT diagnosed by the Child Protective Team. Differential scores accounted for words overrepresented in AHT patient vs. control notes. Sentiment scores were reflective of note positivity/negativity and subjectivity scores accounted for note subjectivity/objectivity. The composite scores reflected the patient's differential score modified by the subjectivity score. Composite, sentiment, and subjectivity scores combined with demographic information trained a Random Forest (RF) machine learning algorithm to predict AHT. RESULTS: Final composite scores with demographic information were highly associated with AHT in a test dataset. The control group included 587 patients and the test group included 193 patients. Combining composite scores with demographic information into the RF model improved AHT classification area under the curve (AUC) from 0.68 to 0.78, with an overall accuracy of 84%. Feature importance analysis of our RF model revealed that composite score, sentiment, age, and subjectivity were the most impactful predictors of AHT. The sentiment was not significantly different between control and AHT notes (p = 0.87), while subjectivity trended higher for AHT notes (p = 0.081). CONCLUSION: We conclude that a machine learning algorithm can recognize patterns within free-text notes and demographic information that aid in AHT detection in children. LEVEL OF EVIDENCE: III.

Characterizing the role of Phlda3 in the development of acute toxicity and malignant transformation of hematopoietic cells induced by total-body irradiation in mice.

The tumor suppressor p53 is a transcriptional factor that plays a crucial role in controlling acute toxicity and long-term malignant transformation of hematopoietic cells induced by genotoxic stress such as ionizing radiation. Among all transcriptional targets of p53, one gene that is robustly induced by radiation is the pleckstrin homology domain-only protein Phlda3. However, the role that Phlda3 plays in regulating the response of hematopoietic cells to radiation is unknown. Here, using isogenic cell lines and genetically engineered mouse models, we showed that radiation induces Phlda3 in human leukemia cells and mouse normal hematopoietic cells in a p53-dependent manner. However, deletion of the Phlda3 gene did not ameliorate radiation-induced acute hematologic toxicity. In addition, distinct from mice that lose p53, loss of Phlda3 did not alter the latency and incidence of radiation-induced thymic lymphoma in mice. Remarkably, whole-exome sequencing data showed that lymphomas in irradiated Phlda3+/+ mice harbor a significantly higher number of single nucleotide variants (SNVs) and indels compared to lymphomas in irradiated Phlda3+/- and Phlda3-/- littermates. Together, our results indicate that although deletion of Phlda3 does not accelerate the development of radiation-induced thymic lymphoma, fewer SNVs and indels are necessary to initiate lymphomagenesis after radiation exposure when Phlda3 is silenced.

Monitoring Sarcoma Response to Immune Checkpoint Inhibition and Local Cryotherapy with Circulating Tumor DNA Analysis.

PURPOSE: Immune checkpoint inhibition has led to promising responses in soft tissue sarcomas (STS), but the majority of patients do not respond and biomarkers of response will be crucial. Local ablative therapies may augment systemic responses to immunotherapy. We evaluated circulating tumor DNA (ctDNA) as a biomarker of response in patients treated on a trial combining immunotherapy with local cryotherapy for advanced STS. PATIENTS AND METHODS: We enrolled 30 patients with unresectable or metastatic STS to a phase II clinical trial. Patients received ipilimumab and nivolumab for four doses followed by nivolumab alone with cryoablation performed between cycles 1 and 2. The primary endpoint was objective response rate (ORR) by 14 weeks. Personalized ctDNA analysis using bespoke panels was performed on blood samples collected prior to each immunotherapy cycle. RESULTS: ctDNA was detected in at least one sample for 96% of patients. Pretreatment ctDNA allele fraction was negatively associated with treatment response, progression-free survival (PFS), and overall survival (OS). ctDNA increased in 90% of patients from pretreatment to postcryotherapy, and patients with a subsequent decrease in ctDNA or undetectable ctDNA after cryotherapy had significantly better PFS. Of the 27 evaluable patients, the ORR was 4% by RECIST and 11% by irRECIST. Median PFS and OS were 2.7 and 12.0 months, respectively. No new safety signals were observed. CONCLUSIONS: ctDNA represents a promising biomarker for monitoring response to treatment in patients with advanced STS, warranting future prospective studies. Combining cryotherapy and immune checkpoint inhibitors did not increase the response rate of STS to immunotherapy.

Germline modifiers of the tumor immune microenvironment implicate drivers of cancer risk and immunotherapy response.

With the continued promise of immunotherapy for treating cancer, understanding how host genetics contributes to the tumor immune microenvironment (TIME) is essential to tailoring cancer screening and treatment strategies. Here, we study 1084 eQTLs affecting the TIME found through analysis of The Cancer Genome Atlas and literature curation. These TIME eQTLs are enriched in areas of active transcription, and associate with gene expression in specific immune cell subsets, such as macrophages and dendritic cells. Polygenic score models built with TIME eQTLs reproducibly stratify cancer risk, survival and immune checkpoint blockade (ICB) response across independent cohorts. To assess whether an eQTL-informed approach could reveal potential cancer immunotherapy targets, we inhibit CTSS, a gene implicated by cancer risk and ICB response-associated polygenic models; CTSS inhibition results in slowed tumor growth and extended survival in vivo. These results validate the potential of integrating germline variation and TIME characteristics for uncovering potential targets for immunotherapy.

Differential regulation of TNFα and IL-6 expression contributes to immune evasion in prostate cancer.

BACKGROUND: The role of the inflammatory milieu in prostate cancer progression is not well understood. Differences in inflammatory signaling between localized and metastatic disease may point to opportunities for early intervention. METHODS: We modeled PCa disease progression by analyzing RNA-seq of localized vs. metastatic patient samples, followed by CIBERSORTx to assess their immune cell populations. The VHA CDW registry of PCa patients was analyzed for anti-TNF clinical outcomes. RESULTS: We observed statistically significant opposing patterns of IL-6 and TNFα expression between localized and metastatic disease. IL-6 was robustly expressed in localized disease and downregulated in metastatic disease. The reverse was observed with TNFα expression. Metastatic disease was also characterized by downregulation of adhesion molecule E-selectin, matrix metalloproteinase ADAMTS-4 and a shift to M2 macrophages whereas localized disease demonstrated a preponderance of M1 macrophages. Treatment with anti-TNF agents was associated with earlier stage disease at diagnosis. CONCLUSIONS: Our data points to clearly different inflammatory contexts between localized and metastatic prostate cancer. Primary localized disease demonstrates local inflammation and adaptive immunity, whereas metastases are characterized by immune cold microenvironments and a shift towards resolution of inflammation and tissue repair. Therapies that interfere with these inflammatory networks may offer opportunities for early intervention in monotherapy or in combination with immunotherapies and anti-angiogenic approaches.

Germline genetic biomarkers to stratify patients for personalized radiation treatment.

BACKGROUND: Precision medicine incorporating genetic profiling is becoming a standard of care in medical oncology. However, in the field of radiation oncology there is limited use of genetic profiling and the impact of germline genetic biomarkers on radiosensitivity, radioresistance, or patient outcomes after radiation therapy is poorly understood. In HNSCC, the toxicity associated with treatment can cause delays or early cessation which has been associated with worse outcomes. Identifying potential biomarkers which can help predict toxicity, as well as response to treatment, is of significant interest. METHODS: Patients with HNSCC who received RT and underwent next generation sequencing of somatic tumor samples, transcriptome RNA-seq with matched normal tissue samples were included. Patients were then grouped by propensity towards increased late vs. early toxicity (Group A) and those without (Group B), assessed by CTCAE v5.0. The groups were then analyzed for association of specific germline variants with toxicity and clinical outcomes. RESULTS: In this study we analyzed 37 patients for correlation between germline variants and toxicity. We observed that TSC2, HLA-A, TET2, GEN1, NCOR2 and other germline variants were significantly associated with long term toxicities. 34 HNSCC patients treated with curative intent were evaluated for clinical outcomes. Group A had significantly improved overall survival as well as improved rates of locoregional recurrence and metastatic disease. Specific variants associated with improved clinical outcomes included TSC2, FANCD2, and PPP1R15A, while the HLA-A and GEN1 variants were not correlated with survival or recurrence. A group of five HLA-DMA/HLA-DMB variants was only found in Group B and was associated with a higher risk of locoregional recurrence. CONCLUSIONS: This study indicates that germline genetic biomarkers may have utility in predicting toxicity and outcomes after radiation therapy and deserve further investigation in precision radiation medicine approaches.