Immunotherapy in Sarcoma – Current Data and Promising Strategies

Soft tissue sarcomas are rare cancers affecting muscle, fat, and other connective tissues. Their rarity can make it difficult to recruit enough patients for clinical trials, contributing to the lack of substantial treatment advances over the past 30 years.

Standard treatment for soft tissue sarcoma varies according to factors such as tumor type, stage, location, and the patient's general health. Treatment generally consists of combinations of surgery, radiotherapy, and chemotherapy. Surgery is often the first choice, aiming to remove the tumor completely. Radiotherapy is used before or after surgery to shrink tumor cells or destroy those remaining. Chemotherapy is administered systemically to target microscopic residual tumor cells. However, approximately 50% of patients with high-risk sarcoma experience recurrence or metastasis after treatment. In recent years, innovative approaches such as immunotherapy and targeted therapies have also begun to play an important role in sarcoma treatment.

Important developments in sarcoma in recent years include:

• Tazemetostat Received FDA Approval for Advanced Epithelioid Sarcoma in January 2020
• Atezolizumab Immunotherapy Received FDA Approval for Alveolar Soft Part Sarcoma in December 2022
• A Milestone in Synovial Sarcoma – The First T-Cell Therapy on the Path to FDA Approval

The following article summarizes notable recent sarcoma immunotherapy research presented at ASCO 2024 and is a translation of the article published in the ASCO Educational Book on May 23, 2024, entitled "Immunotherapy in Sarcoma: Current Data and Promising Strategies".

Immunotherapies in Sarcoma

Traditionally, sarcomas, with low tumor mutational burden (TMB) , reduced T-cell infiltration , and high levels of H1F1α, macrophages, and neutrophils within an immunosuppressive tumor microenvironment (TME) , have been regarded as immunologically silent tumors .

However, research has shown that a subset of sarcomas have high TMB, PDL-1 expression, CD8+ T cells, or tertiary lymphoid structures (TLS), making them immunologically "hot" and sensitive to immunotherapy.

In this article, we review the current evidence for immunotherapy in bone sarcomas (BS) and soft tissue sarcomas (STS), including immune checkpoint inhibitors (ICI), adaptive cellular therapies such as engineered T-cell therapies, chimeric antigen receptor (CAR) T-cell therapies, tumor-infiltrating lymphocytes (TILs), and cancer vaccines, as well as biomarkers of response.

Key Points to Remember

• Sarcoma is a complex group of cancers with many subtypes, each with unique biology and an immune microenvironment. A histology-specific approach is therefore needed in immuno-oncology drug development.
• Checkpoint blockade monotherapy has shown limited effectiveness in most sarcoma subtypes, highlighting the need for alternative approaches and predictive biomarkers.
• Combination strategies involving chemotherapy, radiotherapy, and tyrosine kinase inhibitors, along with new agents, are being developed, but none has yet demonstrated practice-changing evidence.
• Adaptive cell therapy using T-cell receptors targeting cancer-specific antigens (MAGEA4) is most effective in synovial sarcoma. There is a need to broaden its applicability by overcoming human leukocyte antigen restrictions, identify new targets and approaches such as chimeric antigen receptors, and improve durability.
• Tertiary lymphoid structures have emerged as promising predictive biomarkers, at least for checkpoint blockade monotherapy. However, they have a low prevalence in sarcomas, and new therapeutics and more broadly applicable predictive biomarkers are needed.

Immune Checkpoint Inhibitors

Immune checkpoint inhibitors (ICIs) enable the immune system to fight cancer cells more effectively by regulating signals that suppress T cells. These drugs target inhibitory signals affecting T cells, such as CTLA-4 and PD-1/PD-L1. ICIs have been effective both alone as monotherapy and with other treatments as combination therapy. Accordingly, the United States Food and Drug Administration (FDA) has approved these treatments for more than 50 solid tumors, mismatch repair (MMR)-deficient tumors, and tumors with high tumor mutational burden (TMB).

Several clinical studies have demonstrated some success with ICIs in soft tissue sarcomas (STS). The SARC028 phase II trial was the first to show pembrolizumab (a PD-1 inhibitor) activity in patients with either STS (undifferentiated pleomorphic sarcoma [UPS], dedifferentiated liposarcoma [DDLPS], leiomyosarcoma [LMS], synovial sarcoma [SS]) or bone sarcomas (BS) (Ewing's Sarcoma [ES], osteosarcoma [OS], chondrosarcoma [CS]). The objective response rate (ORR) in STS was 18%, with the best responses in UPS (4 of 10 patients) and DDLPS (2 of 10 patients). The ORR in BS was a disappointing 5%. The SARC028 expansion cohort analysis confirmed pembrolizumab's clinical activity in UPS, with an ORR of 23%, while a lower response rate of 10% was observed in DDLPS.

The AcSE phase II trial demonstrated heterogeneous responses to pembrolizumab in rare sarcoma subtypes. Partial responses (PR) of 50% and 25% were observed in alveolar soft part sarcoma (ASPS) and SMARCA4-deficient sarcoma or malignant rhabdoid tumor, respectively. Two complete responses (CR) were observed in patients with ASPS and epithelioid sarcoma.

However, ICIs have produced poor results in BS. SARC28 had an ORR of 5% in the ES, OS, and CS subtypes. The PROMO trial reproduced disappointing results for pembrolizumab in 12 patients with osteosarcoma (OS) and was stopped early. Similarly, a pediatric phase I/II pembrolizumab trial involving melanoma, PD-L1-positive solid tumors, primarily ES and OS, and lymphoma showed PR in 3 of 33 sarcoma patients.

Investigations of other ICI monotherapies, such as the PD-1 inhibitor nivolumab or the CTLA-4 inhibitor ipilimumab, have been less productive. The nivolumab monotherapy arm of the Alliance trial recorded a response rate of only 5% and was not pursued further.

An observational study of sarcoma patients treated with ICIs identified enriched responses in ASPS, leading to prospective studies demonstrating the benefit of PD-1/PD-L1 inhibition in this histologic subtype. Atezolizumab produced excellent results in ASPS, with durable responses in one third of patients, leading to FDA approval in this rare subtype. However, this success was not reproduced in the broader sarcoma population in the iMATRIX phase I/II atezolizumab trial. Among 40 pediatric patients with BS + STS, only six achieved stable disease, and no objective responses were observed.

Combination Immunotherapy: Combination Strategies

Because ICI monotherapy has largely proved unsuccessful for most sarcoma patients, dual checkpoint blockade and other combinations were evaluated in the phase II Alliance trial. Patients treated with nivolumab and ipilimumab had an ORR of 16%, with responses in LMS, myxofibrosarcoma (MFS), UPS, and angiosarcoma. The expansion cohort showed an ORR of 14% for dual checkpoint blockade in UPS and DDLPS. The phase II DART trial confirmed nivolumab/ipilimumab activity in angiosarcoma, with an ORR of 25%. A phase II durvalumab + tremelimumab study showed similar histology-specific patterns of activity, with the best responses in ASPS, as in other studies. Results of the phase III RAR-Immune trial investigating nivolumab and ipilimumab in sarcomas are eagerly awaited (ClinicalTrials.gov identifier: NCT04741438).

A recent pilot study of bempegaldesleukin plus nivolumab in metastatic sarcoma showed the highest ORR in angiosarcoma + UPS. CD8+ T-cell infiltrates and PD-1 expression were associated with improved ORR.

This was echoed in a retrospective analysis of 88 patients with metastatic STS who received pembrolizumab, nivolumab, ipilimumab, or their combinations, showing greater immunotherapy responses in UPS and LMS.

Combination with Local Treatment

Local radiotherapy (RT) can generate tumor-associated antigens (TAAs), increasing the immunogenicity of low-TMB tumors and their sensitivity to ICIs in a T-cell-dependent manner. Neoadjuvant trials have shown highly promising results with RT plus ICIs, with substantial activity in UPS.

Isolated limb infusion (ILI) is the local delivery of high-dose chemotherapy to treat STS of the extremities. A successful case report describing tumor regression after ILI plus pembrolizumab in two patients with recurrent MFS led to a phase II study investigating this combination (ClinicalTrials.gov identifier: NCT04332874).

Talimogene laherparepvec (T-VEC), an oncolytic viral immunotherapy delivered by intratumoral injection, is approved for metastatic melanoma. A phase II study combining it with pembrolizumab in advanced sarcoma showed an ORR of 35% and a mean response duration of 56.1 weeks. Responses occurred in patients who had previously received checkpoint blockade.

Combination with Systemic Therapies

Evidence suggests that combining ICIs with chemotherapy improves effectiveness in STS. Two clinical studies evaluated the synergistic activity of pembrolizumab plus doxorubicin, an anthracycline chemotherapy that is a treatment backbone for many sarcoma histotypes, with promising results. The Pollack study showed an ORR of 19%, similar to pembrolizumab or doxorubicin alone, but favorable median progression-free survival (mPFS) and median overall survival (mOS) of 8.1 months and 27 months, respectively. The Livingstone study showed a higher ORR of 36.7%, but improvements in mPFS and mOS remained modest (5.7 months + 17 months). Responses occurred in UPS (4 of 4) and LMS (4 of 10).

The phase II PEMBROSARC trial evaluated pembrolizumab plus metronomic cyclophosphamide in gastrointestinal stromal tumors (GIST), LMS, UPS, and other sarcomas. Its latest update showed PR as the best response in nine patients: five with DDLPS, three with epithelioid sarcoma, and one with LMS. Only 6% of the 15 osteosarcoma patients in the trial (1 of 15) achieved PR. No association was found between PD-L1 expression and response in this cohort.

A retrospective analysis showed that trabectedin plus nivolumab was safe and effective in STS, subsequently confirmed by the recent NiTraSarc study. Trabectedin has also shown clinical activity with ICIs, including ipilimumab/nivolumab in the SAINT trial (disease control rate [DCR], 87.3%; ORR, 19.5%). Other combinations with avelumab (ORR, 13%) or durvalumab (ORR, 7%) did not show synergistic activity with trabectedin. The combinations were safe, with tolerable toxicity profiles. Most treatment-related grade 3-5 events were attributed to trabectedin, including two fatal cases of febrile neutropenia.

ICI plus tyrosine kinase inhibitor (TKI) combinations have also been evaluated. In the phase II ImmunoSarc trial, patients with advanced sarcoma received sunitinib and nivolumab, with tolerable toxicity, an ORR of 22%, and 100% overall survival at 18 months (PR in ASPS [n = 2], angiosarcoma [n = 1], extraskeletal myxoid CS [n = 1], and SS [n = 1]). One patient with angiosarcoma achieved CR.

Pembrolizumab plus axitinib in STS showed clinical benefit in 73% of patients with ASPS, with a PR rate of 54.5% and an SD rate of 18%, leading to an expansion cohort for this histotype. In patients without ASPS, the ORR was only 9.5%, while the clinical benefit rate was 42.9%. A pilot study of pembrolizumab plus lenvatinib also showed promising results, with PR in OS, malignant peripheral nerve sheath tumor, angiosarcoma, and SS.

A phase II nivolumab plus cabozantinib study in advanced angiosarcoma (Alliance A091902) showed substantial responses, with an ORR of 59%, including two CRs; mOS was not reached. These histotype-specific responses were highlighted by studies in GIST (dasatinib + ipilimumab) and BS, in which TKI plus ICI combinations provided no substantial benefit. No immunomodulatory effect of treatment was observed in those studies.

Interestingly, a meta-analysis of STS found higher responses with TKI plus anti–CTLA-4 in ASPS, while the best responses occurred in UPS with chemotherapy plus anti–PD-1 and anti–CTLA-4. Overall, grade 3-5 toxicity ranged from 7% to 76%. Chemotherapy combinations caused substantial hematologic toxicity (absolute risk [AR], 0.74 [95% CI, 0.60 to 0.85]), whereas TKI combinations were more manageable (AR, 0.18 [95% CI, 0.11 to 0.27]).

Adaptive Cell Therapy

Engineered T-Cell Receptor Therapy

T-cell receptor (TCR)-based adaptive cell therapy (ACT) is one way of targeting TAAs. Sarcomas express higher-than-normal levels of TAAs known as cancer-testis antigens (CTAs), particularly SS and myxoid/round cell liposarcoma (MRCLS). CTAs expressed by sarcomas include NY-ESO-1, MAGE, GAGE, and fetal acetylcholine receptors.

In 2011, the first clinical trial of engineered TCR therapy for sarcoma evaluated autologous NY-ESO-1–specific TCRs with high-dose interleukin-2 (IL-2) in HLA-A*021-positive patients with refractory SS. The initial analysis showed RECIST PR in four of six patients.

Long-term follow-up found that 11 of 18 SS patients (61%) receiving anti-NY-ESO-1 T cells responded, including one complete response (CR).

A subsequent phase I study of SPEAR TCRs targeting NY-ESO-1 treated HLA-A*021-positive patients with advanced SS after lymphodepletion with fludarabine and cyclophosphamide. An ORR of 50% (6 of 12 patients) and one CR were observed. Correlative analyses showed expansion and persistence of polyfunctional SPEAR T cells (Ny-ESO-1c259T) that self-renewed in vivo for at least 6 months after infusion.

Patients receiving dose-reduced lymphodepletion regimens had lower ORRs. Fludarabine also appears necessary, as in hematologic malignancies. NY-ESO-1–specific TCR therapy has also been investigated in MRCLS, showing ORRs of 20% and 40% in two cohorts receiving low-dose and standard-dose chemotherapy, respectively.

ADP-A2M4 SPEAR TCRs directed against MAGE-A4 showed similar promise. An initial phase I study in advanced solid tumors found an ORR of 25%, with PR in 44% of patients with SS. The subsequent Spearhead study in advanced SS or MRCLS showed an ORR of 39%.

Chimeric Antigen Receptor T-Cell Therapy

CAR T cells have emerged as effective treatments in several hematologic malignancies, resulting in FDA approvals. Progress in solid tumors has been challenging, but signals of activity have been found in sarcoma. Taking advantage of low-level human epidermal growth factor receptor 2 (HER2) expression in sarcomas, a phase I/II trial using HER2-specific CAR T cells, primarily in osteosarcoma (OS), found a median overall survival (OS) of 10.3 months. HER2-CAR T cells were detected in some tumor samples tested at resection, and one patient had 90% necrosis. Further evaluation in 10 sarcoma patients showed one-year overall survival of 60%, with one CR in OS and another CR observed. Other CAR T-cell trial targets recruiting in sarcoma include GD2 (ClinicalTrials.gov identifiers: NCT03373097, NCT05080790), which has shown promise in diffuse midline gliomas and glioblastoma. The EGFR+/– CD19 STRIvE-01 study showed mixed responses in 3 of 11 patients at early dose levels.

Tumor-Infiltrating Lymphocyte Therapies

TIL therapy is now FDA-approved for melanoma. Following favorable preclinical research, a phase I study is underway to assess the feasibility of TIL therapy in adolescent and young adult patients with STS (ClinicalTrials.gov identifier: NCT04052334).

Targeting CTAs with Cancer Vaccines

Cancer vaccines aim to trigger an adaptive antitumor immune response by stimulating both cellular and humoral immunity. A patient with Ewing sarcoma (ES) was treated with intratumoral Pexa-Vec in the pubis and achieved stable disease for 22 days after treatment, although disease progressed elsewhere.

In one patient, the NY-ESO-1–expressing lentivirus LV305 induced T-cell and antibody responses, but the ORR in sarcoma patients was 4.2%. A NY-ESO-1 dendritic cell vaccine administered with autologous NY-ESO-1–directed T cells and ipilimumab showed a 90-day ORR of 20% (2 of 10 patients, including one CR). Adding ipilimumab provided no clear clinical benefit.

The heterologous vaccine CMB305 sought to improve response rates by adding a TLR-4 agonist. Phase 1b evidence suggested that it was a useful platform for SS and MRCLS, with a DCR of 61.9%. However, progression-free and overall survival results in the phase II CMB305 study with atezolizumab were similar to published results for standard chemotherapy.

Experience with NY-ESO-1 vaccines in STS has been disappointing, with the best ORRs equivalent to standard-of-care chemotherapy. One patient had a durable response, but also received ACT, raising the question of whether the excellent response was due to vaccination rather than ACT.

An important subset of STS is driven by pathognomonic fusion events. These generate immunogenic chimeric proteins (TAAs) that often function as oncogenic drivers. Targeting these fusion-derived neoantigens has proved successful in other tumor types. In STS, targeting fusion-derived TAAs using a peptide-loaded dendritic cell vaccine approach has not yet produced clinically meaningful responses.

Targeting the Tumor Necrosis Factor-α Family Pathway

Identifying tumor necrosis factor (TNF)-α ligand pathways as a means of selectively inducing apoptosis in tumor cells generated enthusiasm for compounds targeting receptors for family members. However, early hepatotoxicity and a lack of potency have largely impeded these efforts. Nevertheless, a third-generation functional mimic of TNF-related apoptosis-inducing ligand (TRAIL), which binds death receptor 5 (DR5) and triggers its clustering, produced exciting early data in the KS cohort of a phase I study. INBRX-109 yielded a DCR of 87%, with durability exceeding 180 days. PFS was 7.6 months, and the 6-month PFS rate was 58%. These phase I data led to an ongoing registrational study in conventional KS (ClinicalTrials.gov identifier: NCT03715933).

A separate ongoing study (ClinicalTrials.gov identifier: NCT03715933) is also evaluating the drug in two other sarcoma subtypes, ES and SDH-deficient GIST. Earlier ES studies identified DR5 on the surface of primary tissue samples, and DR5 was identified in GIST cell lines. The study combines INBRX-109 with temozolomide and irinotecan in refractory ES, while patients with SDH-deficient GIST receive it alone.

Modifying the Tumor Microenvironment

The tumor microenvironment (TME) creates barriers to immune recognition and activation through several mechanisms. New studies targeting this immunosuppressive environment attack different signaling molecules that suppress immunity. Tumor-associated macrophages (TAMs) differentiated from myeloid-derived suppressor cells (MDSCs) secrete angiogenic factors that promote cancer cell growth and improve oxygen and nutrient delivery, while also secreting immunosuppressive cytokines. Engagement of colony-stimulating factor 1 receptor (CSF1R) enables precursor myeloid cells within the TME to become immunosuppressive M2 macrophages. Strategies to block CSF1R signaling include small-molecule intracellular inhibitors of receptor signaling and extracellular inhibitory antibodies against the CSF1 ligand or the receptor itself. A phase I trial investigating the small-molecule inhibitor DCC-3014 plus avelumab in metastatic sarcoma found a decrease in circulating MDSCs and stable disease in three of 13 patients (ClinicalTrials.gov identifier: NCT04242238).

A second novel target supporting tumor growth in the TME is CD73. Sarcomas express high levels of CD73 on their cell surfaces, and this widely distributed nucleotidase converts adenosine monophosphate into extracellular adenosine. Adenosine signaling can increase the metastatic potential of tumor cells, while binding to receptors on immune cells downregulates T-cell, natural killer (NK)-cell, and macrophage function. Clinical trials are attempting to overcome the inhibitory TME by interfering with adenosine. A phase II study is evaluating oleclumab, a targeted CD73 inhibitor, with durvalumab across sarcoma subtypes. A second study is evaluating the oral adenosine antagonist etrumadenant plus the PD-1 inhibitor zimberelimab in DDLPS (ClinicalTrials.gov identifier: NCT05886634).

Another widely expressed transmembrane protein contributing to immunosuppression in the TME is CD47. It binds several proteins, including members of the signal regulatory protein family, integrins, and thrombospondin. Its interaction through adaptor ligands on macrophages prevents cellular phagocytosis. By expressing CD47, tumor cells reduce macrophage activity in the TME. CD47 is expressed in OS and various STS. In preclinical models, CD47 blockade increased cancer cell phagocytosis and reduced metastatic disease. A phase I/II trial showed clinical activity with ontorpacept, a soluble fusion decoy receptor, plus doxorubicin in advanced LMS, achieving a DCR of 80%.

Biomarkers of Immunotherapy Response in Soft Tissue Sarcoma

Activity Across Sarcoma Histotypes

ICI trials including all STS histotypes have generally been disappointing, with response rates of only about 15%. A meta-analysis of phase II trials found an overall response rate of 15.1% to PD-1/PD-L1 inhibitors. However, substantial differences were observed across histologies. UPS in particular showed stronger responses than LPS, LMS, or SS. The histotype responding best to ICIs (PD-1/PD-L1 and/or CTLA-4 inhibition) was rare ASPS, with response rates exceeding 35%, far higher than other histotypes. Given the generally low response rates among the most common histotypes, however, cross-histotype or histotype-agnostic biomarkers of ICI response are needed.

The Importance of Pan-Cancer Biomarkers for Soft Tissue Sarcoma

Tumor Mutational Burden

TMB measures the number of mutations in a tumor's exome. Higher response rates to PD-1 blockade have been reported in patients with high TMB across many solid tumor types. This study and the subsequent FDA approval use a threshold of 10 mutations per megabase (Mb) to distinguish high-TMB from low-TMB tumors. Most soft tissue sarcomas, however, have very low TMB, with a median below three mutations per Mb. Only 2% are classified as TMB-high, making TMB an impractical biomarker for STS.

Interestingly, some histologies have a higher frequency of TMB-high tumors, particularly angiosarcoma. The Angiosarcoma Project reported higher TMB in head and scalp angiosarcomas than in angiosarcomas at other sites (20.7 mutations/Mb). Two patients with head and scalp angiosarcoma benefited from PD-1 blockade, while three patients with angiosarcoma at other sites had no clinical benefit. Although these numbers are insufficient to draw conclusions about the relationship between TMB and immunotherapy response, they indicate a need for further study.

Mismatch Repair Deficiency and Microsatellite Instability

Mismatch repair deficiency (MMRd) in tumors is associated with higher TMB and may lead to microsatellite instability (MSI). MMRd is associated with higher response rates to PD-1 blockade. However, MMRd and MSI are rare in STS, accounting for only 1% of tumors. Notable exceptions are pleomorphic RMS and embryonal RMS, which have a high frequency of MMRd.

Importantly, MMRd STS have lower TMB than most nonsarcoma MMRd tumors. Consequently, MMRd and MSI have limited applicability for immunotherapy in STS.

PD-1 Expression

PD-L1 expression by cancer or immune cells is considered a potential TME biomarker of response to PD-1/PD-L1 inhibitors and has been approved as a biomarker by the FDA and EMA. Wide variation in PD-L1 expression across STS histotypes and tumors has been reported. Correlative analysis of SARC028 showed stronger PD-L1 expression in both tumor and immune cells among responders to PD-1 blockade. This association has not been reported in OS. Prior treatment, such as RT, may also influence PD-L1 expression. These findings suggest that PD-L1 could be a valuable biomarker for immunotherapy in STS, but further studies are needed.

TME and Response to Immune Checkpoint Inhibitors in STS

The cellular composition of tumors may include not only malignant cells but also fibroblasts, endothelial cells, and almost every type of immune cell. These diverse cell populations interact extensively and are collectively referred to as the TME. Across many cancer types, TME composition has been strongly linked to ICI response or resistance.

B Cells and TLS

In-depth transcriptomic characterization of the TME in soft tissue sarcoma enabled an immune classification based on five Sarcoma Immune Classes (SICs). Classes A, C, and E had distinct profiles, respectively described as immune desert, highly vascularized, and immune high. Immune-high SIC E was associated with longer survival, and B cells were identified as the main determinant of patient survival. One feature of SIC E tumors is the presence of TLS. These are lymph node-like aggregates of lymphocytes in or near the tumor, with mature TLS displaying a functional germinal center. They can perform a broad range of functions, including initiating and activating antitumor immune responses. They have been shown to improve patient survival in many malignancies and can be viewed as part of a local cancer immunity cycle.

SIC E tumors had a much higher response rate to pembrolizumab than other SICs, while no responders were found in immune-low SIC A. TLS have also been reported as potential biomarkers in melanoma and kidney cancer, suggesting broader applicability. A pan-cancer study concluded that TLS with mature germinal centers predicted ICI response independently of PD-L1 expression. To validate these findings, PEMBROSARC included another cohort enrolled on the basis of TLS in their tumors and treated with pembrolizumab plus low-dose cyclophosphamide. An ORR of 30% was reported in the TLS+ cohort, substantially higher than the 2.4% ORR in the overall cohort. In TLS+ tumors, response was associated with lower regulatory T-cell density and greater infiltration by fully differentiated B cells specialized in antibody production. This recalls similar findings in clear cell renal cell carcinoma and non-small cell lung cancer. Consistent with these results, a recent durvalumab plus pazopanib study showed improved PFS with greater B-cell presence.

CD8+ T Cells

The presence of potentially cytotoxic CD8+ T cells has been associated with increased ICI response across cancer types. The SARC028 correlative study reported greater CD8+ T-cell infiltration in responding tumors. Further analysis of this dataset showed a stronger association between CD8+ T cells and response in tumors with high neoantigen burden. This pattern was not seen when combining CD8+ T cells with TMB, suggesting that neoantigen burden may have a greater predictive role.

A study combining the oncolytic immunotherapy T-VEC with the PD-1 inhibitor pembrolizumab reported an association between pretreatment CD8+ T cells at the tumor periphery and response. A study of the IL-2 pathway agonist bempegaldesleukin plus anti–PD-1 nivolumab found a correlation between response, CD8+ T-cell infiltrates, and PD-1 expression, while Hedgehog pathway expression was associated with treatment resistance. Immune-hot tumors enriched in T cells, B cells, and NK cells were more frequent among responders to various ICI combinations. Interestingly, this immune-hot tumor class was also associated with the expression of transposable elements and epigenetic regulators.

Other TME Markers

TAMs are an expanding area of research, with many therapeutic strategies designed to target them. However, their relationship with immunotherapy in STS has been little studied. The SARC028 correlative study reported increased macrophage presence in tumors responding to pembrolizumab. Larger studies are needed to assess the potential of tumor-associated macrophages as immunotherapy biomarkers.

A recent study showed a correlation between immune-related adverse events and survival in sarcoma patients treated with ICIs. Tumors in patients experiencing immune-related adverse events had greater infiltration by dendritic cells, CD8+ T cells, and regulatory T cells. Other inflammatory indices, such as a low lymphocyte-to-monocyte ratio, have been highlighted as prognostic or treatment-response indicators.

Conclusions and Future Directions

Bone sarcomas (BS) and soft tissue sarcomas (STS) remain devastating diagnoses with poor prognoses in advanced disease. Although treatment options are expanding, durable responses to checkpoint blockade remain difficult to achieve. Recent efforts have shown that some STS are indeed immunogenic and that immunotherapy may play a role in their management. Overall ICI response rates in STS are not high, but histology-specific activity exists. More detailed immune characterization of these ICI-sensitive tumors will help guide treatment strategies. The differing responses highlight the need for histology-specific approaches, which should be incorporated into clinical trial design and correlative studies to identify predictive biomarkers.

Results in BS have been more disappointing, with checkpoint blockade largely ineffective. Alternative immunotherapeutic options need to be investigated.

Given the TME of sarcomas, most histologic subtypes will probably require a multimodal approach to induce effective antitumor immunity. Accordingly, combinations with RT or other systemic treatments have improved ORRs.

Similarly, ACT approaches such as TCR-based treatment targeting CTAs have shown impressive response rates in advanced SS and MRCLPS, but most responders ultimately experience recurrence. Understanding primary and acquired resistance mechanisms in these patients will be important for improving treatment strategies. Broader use of TCR-based treatments through technologies that remove HLA restrictions, along with broader identification of TAAs, would also be advantageous.

Given the excellent responses achieved with engineered TCR approaches in STS and the disappointing results of dendritic cell vaccines targeting the same antigen, the role of TCR-based therapy targeting fusion-derived neoantigens warrants further investigation.

How Can This Information Be Used in Clinical Practice?

Currently, atezolizumab for ASPS remains the only FDA approval for checkpoint blockade in sarcoma. Access to pembrolizumab or nivolumab through the National Comprehensive Cancer Network (NCCN) compendium remains an option in the United States for angiosarcoma, UPS, and DDLPS, but identifying response and resistance biomarkers is urgently needed even in these subtypes. This emphasizes the need to investigate these agents within clinical trials. More international studies would also improve global access. Because checkpoint blockade monotherapy has activity in only a few histologies, combination strategies will take priority, with careful consideration of their risk-benefit balance. Adding ipilimumab does not justify the increased toxicities observed in the Alliance expansion cohorts in UPS/DDLPS.

Clinical trials should continue in histology-specific cohorts, and tissue collection in these studies is essential. Although TLS are the most promising biomarker to date, TLS selection yielded a response rate of 30% to checkpoint blockade. Ultimately, a panel of biomarkers will be needed to guide clinicians.

New checkpoints targeting lymphocyte activation gene 3 (LAG-3) may be considered in sarcoma. Identifying drugs that can modulate regulatory T cells and myeloid-derived suppressor cells is also important. Few compounds can currently target these pathways effectively. New bispecific fusion proteins targeting immuno-oncology pathways that downregulate CD80, upregulate IL2, and activate NK cells are in active clinical trials.

In adaptive cell therapy, a TCR targeting MAGEA-4 (afami-cel) is under FDA review. It could become the second ACT approved for solid tumors, highlighting progress while emphasizing the need to extend applicability to other HLA subtypes. Identifying resistance mechanisms, such as loss of antigen expression or reduced antigen presentation machinery, will also advance the field. Following lifileucel approval in ICI-refractory melanoma, this is an intriguing avenue to consider in sarcoma as well.

Although some progress has been made, many questions remain unanswered. Multi-institutional collaborations, in both clinical trials and correlative studies, are critical for the advances urgently needed in this field.