Making progress in a rare disease: emerging therapeutics in soft tissue sarcomas

Introduction

Soft tissue sarcomas (STS) represent a heterogeneous group of rare mesenchymal neoplasms with varied clinical presentations and behavior. In 2018, 13,040 new cases will be diagnosed¹. Over 50 histologic subtypes have been identified, with the most commonly represented subtypes including gastrointestinal stromal tumors (GISTs), undifferentiated pleomorphic sarcomas, liposarcomas, and leiomyosarcomas. They can occur anywhere in the body and have widely variable clinical courses ranging from indolent to highly aggressive. In general, the most common sites of presentation are the extremities, but histologic subtypes demonstrate predilections for various parts of the body. The heterogeneity of histologic subtypes creates significant obstacles in identifying unifying treatment strategies across the many subtypes, since the rarity of these tumors has posed limitations in adequately powering clinical trials. The vast differences in biologic drivers of disease and clinical behavior have posed tremendous challenges to the development of effective treatment options for these tumors. Increasingly, clinical trials for STS have become more focused on specific histologic subtypes or tissue-agnostic molecular drivers of disease. The emergence of rationally designed studies and novel experimental therapeutics promises hope for expanded treatment options for patients with unresectable and metastatic STS. Here, we describe the most promising therapies to affect sarcoma management in the near future.

Gastrointestinal stromal tumors

GISTs represent one of the most commonly diagnosed STS. Roughly 5,000 new cases are diagnosed in the United States annually. These tumors can occur anywhere along the GI tract and are thought to be derived from the interstitial cells of Cajal, the pacemakers of the gastrointestinal tract. GISTs are most commonly found in the stomach (60%), small intestine (30–35%), and other locations in the GI tract (<10%)². Gain-of-function mutations in the growth factor receptor genes for KIT (CD117) and platelet-derived growth factor α (PDGFRA) drive tumor development in 85–90% of GISTs to promote constitutive activation of growth factor signaling pathways that cause uncontrolled cell proliferation.

GISTs are considered unresponsive to traditional cytotoxic chemotherapies and radiation. Prior to 2000, no effective systemic therapeutic options existed for patients with unresectable or metastatic disease. In 2000, the first report of response to the tyrosine kinase inhibitor (TKI) imatinib mesylate was published, and this was soon followed by an open-label randomized multi-center clinical trial in 147 patients that demonstrated partial response in 53.7% and stable disease in 27.9% of patients^3,4. Multiple subsequent trials have now firmly established imatinib as the first-line systemic therapy for the management of GISTs. Two other TKIs have since been approved by the US Food and Drug Administration (FDA) for the treatment of GISTs, including sunitinib and regorafenib. Both of these have limited response rates (14% for sunitinib and 4.5% for regorafenib) but did demonstrate improved progression-free survival (PFS) compared to placebo in phase III trials^5,6. While imatinib re-introduction after progression on approved agents has demonstrated an improved PFS over placebo, the duration of benefit is transient (imatinib versus placebo 1.8 versus 0.9 months; p=0.009)⁷. Other TKIs, such as pazopanib, sorafenib, and nilotinib, have demonstrated marginal activity but may be considered as “off-label” options when clinical trial options are unavailable^8–11.

Mutational status of KIT and PDGFRA is crucial to predicting response rates and durability of response to imatinib therapy. The most common mutations found in untreated GISTs are in the juxtamembrane region, specifically in exon 11 (70%) and exon 9 (15%), with up to another 10% of primary mutations being found in PDGFRA¹². KIT exon 11 mutations and insertions portend favorable prognosis, while deletions in exon 11 and exon 9 portend decreased durability of response and shorter survival rates¹³. The most common PDGFRA mutation is the D842V substitution in the activation loop of PDGFRA. Tumors harboring this mutation are refractory to nearly all available treatment options and are associated with poor outcomes with a short PFS of only 2.8 months with imatinib¹⁴.

The major challenge facing the treatment of GISTs remains resistance to currently available targeted therapies. Imatinib functions by binding the ATP-binding pocket for KIT and PDGFRA to prevent the activation of downstream signaling. Development of mutations in the ATP-binding pocket or activation loop decreases the drug’s binding affinity, leading to TKI resistance. The most commonly found secondary resistance mutations are encoded in exons 13 and 14 of the ATP-binding pocket and in exons 17 and 18 of the activation loop. About 14% of tumors have primary resistance to imatinib (progression within the first 6 months of treatment), while in 40–50% of tumors secondary resistance due to acquired mutations in KIT or PDGFRA develops after about 2 years on imatinib treatment^3,12. Activation loop mutations accumulate with sunitinib treatment. For sunitinib and regorafenib, resistance typically develops within approximately 6 months of therapy^5,6. Selection of agents based on the initiating and acquired resistance mutations in KIT and PDGFRA will hopefully expand systemic options beyond the three currently approved TKIs.

Currently, no approved treatment options are available for the highly resistant PDGFRA D842V exon 18 activation loop mutation. However, a number of promising therapeutic agents have emerged. Crenolanib has been specifically developed to target this mutation and demonstrated a clinical benefit rate of 31% in 16 patients¹⁵. A randomized phase III study is currently underway evaluating the efficacy of crenolanib compared to placebo in patients with PDGFRA D842V tumors (CrenoGIST: NCT02847429)¹⁶.

BLU-285, also known as avapritinib, and DCC-2618 are highly potent and selective inhibitors of mutant KIT and PDGFRA, with demonstrated activity against PDGFRA D842V mutations along with secondary KIT resistance mutations. BLU-285 is a selective inhibitor of KIT and PDGFRA activation loop mutants¹⁷. While BLU-285 demonstrated strong activity against clinically relevant single mutations in either the activation loop or the ATP-binding pocket of KIT and PDGFRA, tumor sensitivity was increased in the setting of dual mutants of the juxtamembrane and ATP-binding pocket and protein regions (e.g. exon 11/exon 17). In a phase I study of BLU-285, all 31 patients with D842V PDGFRA GISTs demonstrated a tumor response¹⁸. A phase III study of BLU-285 compared to regorafenib in the third-line setting is planned (VOYAGER: NCT03465722).

DCC-2618 is a pan-KIT and PDGFRA inhibitor with activity against both initiating and acquired resistance mutations (KIT exons 9, 11, 13, 14, 17, and 18 and PDGFRA exon 18). Updated results from a phase I study of 57 heavily pre-treated GIST patients with two or more prior agents were presented at the European Society of Medical Oncology (ESMO) 2017 meeting, revealing a disease control rate of 76% at 12 weeks. DCC-2618 showed partial metabolic response in 69% of patients with KIT or PDGFRA mutants¹⁹. Based on these encouraging results, a randomized phase III pivotal study in GIST patients for treatment in the fourth-line setting is ongoing (INVICTUS: NCT03353753). A second randomized phase III study in the second line versus sunitinib is planned.

Non-gastrointestinal stromal tumor soft tissue sarcomas

Locally aggressive non-malignant soft tissue tumors

Although not classified as malignant given the lack of metastatic potential, certain connective soft tissue tumors cause significant morbidity owing to locally infiltrative disease and mass effect with a high propensity for local recurrence. These include desmoid tumors and tenosynovial giant cell tumors (TGCTs)/pigmented villonodular synovitis (PVNS).

Historically, the management of desmoid tumors included surgical resection, but, given the morbidity and high local recurrence rates, conservative measures, in particular observation, are now increasingly favored. In symptomatic desmoid tumors, new therapies have arisen that target proliferative mechanisms of tumor growth. No FDA-approved therapies exist for desmoid tumors, and the selection of systemic therapy is based, in part, on patient preference and co-morbidities. Efficacy data are limited to small retrospective studies for traditional agents, including non-steroidal anti-inflammatory drugs (NSAIDs), anti-hormonal drugs, and cytotoxic agents such as liposomal doxorubicin. Because of increased expression of c-KIT and PDGFRA/B, TKIs such as imatinib and sorafenib have been investigated as potential therapies. Phase II studies with imatinib demonstrated low response rates from 5–19% with a 6-month PFS of 63%^43–45. Slightly improving upon this, a retrospective study of 26 patients receiving sorafenib showed an objective response in 25% and symptomatic clinical benefit in 70% of patients⁴⁶. This prompted a randomized, double-blind phase III study of sorafenib versus placebo in 87 patients. At interim analysis, this study showed that sorafenib showed a benefit of PFS, which was not reached, compared to 11.3 months for placebo (HR 0.14, 95% CI 0.06–0.33, p<0.0001) and an objective response rate of 33% for sorafenib versus 20% in the placebo group (Alliance A091105: NCT02066181)⁴⁷. A 20% response rate was observed for patients who received placebo, reinforcing observation as an option for these patients given the potential for spontaneous regression.

In TGCTs/PVNS, a translocation involving the colony-stimulating factor (CSF1) gene promotes proliferative inflammation of the synovium through the recruitment of tumor-associated macrophage cells expressing the CSF1 receptor (CSF1R)^48,49. Surgical resection is the mainstay of treatment but has the potential for significant morbidity and high risk for local recurrence. Previously, TGCTs/PVNS had no effective medical therapies. TKIs with activity against the CSF1/CSF1R pathway, such as imatinib and nilotinib, have previously been evaluated with limited objective responses, although disease stabilization was seen in a majority of patients^50,51. The TKI pexidartinib (PLX3397) blocks the CSF1/CSF1R axis, which inhibits tumor-associated macrophage infiltration along with autocrine and paracrine signaling of CSF1 to inhibit tumor cell growth⁵². A marked objective response and durable response in some patients was seen in early studies resulting in the development of the phase III ENLIVEN study (NCT02371369)^53,54. In this study, 120 patients were randomized to either pexidartinib or placebo. A 39% overall response rate (ORR) at week 25 with no further progression at the median 6-month follow-up was seen in responders, which was a vast improvement over prior systemic therapies. Improvements were seen also in a number of functional and symptom scales. In addition, emactuzumab, a monoclonal antibody targeting CSF1R, showed an 86% objective response rate in a phase I study of TGCT/PVNS patients⁵⁵.

Immunotherapy

With the excitement of immune checkpoint blockade in many tumor types, a number of trials have evaluated immunotherapy in the setting of metastatic STS. In general, responses have been limited across a broad range of histologic subtypes. However, selected subtypes of STS appear to demonstrate potential activity. These include undifferentiated pleomorphic sarcomas and dedifferentiated liposarcomas with objective response rates of 40% and 20%, respectively (n=10 for each histologic subtype), with pembrolizumab monotherapy in the SARC028 study⁵⁶. In the ALLIANCE study of nivolumab with or without ipilimumab, nivolumab monotherapy demonstrated a response rate of 5%, while the combination of nivolumab with ipilimumab revealed a response rate of 16%⁵⁷. Combination immune checkpoint strategies continue to be evaluated in multiple histology-specific phase I/II studies. In data presented at the 2018 ASCO annual meeting, pembrolizumab combined with axitinib, a small molecule inhibitor targeting VEGFR, c-KIT, and PDGFR, demonstrated promising activity in STS, particularly in ASPS, where an ORR of 45% was reported with preliminary evidence of durable responses⁵⁸. Case reports and small case series have additionally supported the concept of checkpoint inhibition in ASPS^59,60.

Novel strategies of targeting tumor antigens, such as New York esophageal squamous cell carcinoma 1 (NY-ESO-1), are under active investigation and include adoptive T-cell transfer, chimeric antigen receptor-T cell, and vaccine-based therapies. These are being explored particularly in synovial and myxoid/round cell liposarcoma where >80% of tumors express the antigen (NCT03520959)^61–63.

Neurotrophic tyrosine receptor kinase-targeted therapies

The recent emergence of agents targeting neurotrophic tyrosine receptor kinase (NTRK) has generated significant excitement given case reports of marked responses in NTRK-rearranged sarcomas⁶⁴. These molecularly targeted agents are currently under study in tissue-agnostic phase II basket studies. In the NAVIGATE study, larotrectinib (LOXO-101) demonstrated an 80% ORR in all tumor types, including in 10 of 11 non-GIST STS and 3 of 3 GISTs⁶⁴.

Entrectinib (RDX-101), an oral inhibitor with activity against NTRK, ROS1, and ALK-rearranged tumors, has shown promising activity in early phase clinical trials and is being studied in the phase II STARTRK-2 study (NCT02568267)⁶⁵.

Summary of emerging therapeutics in sarcoma

As we enter a new era of precision medicine and immunotherapy, the future of sarcoma appears poised for the discovery of rationally targeted and histology-specific therapeutic agents. The major obstacles facing sarcoma management remain the rarity and heterogeneity of the disease as well as the potential challenges of recruiting enough patients to a subtype-specific trial to make reasonable conclusions about therapeutic efficacy. With these challenges, developmental therapeutics in sarcoma have smartly shifted to targeting the tumors based on biologic mechanisms. A number of therapeutic agents are poised to change the management for some histologic subtypes (Table 1). Increased understanding of the molecular pathogenesis of histologic subtypes has widened the therapeutic possibilities. The future will entail taking advantage of the unique biology of each histologic sarcoma subtype and identifying relevant biomarkers to drive therapeutic decision making in order to improve outcomes for this unique patient population.

Abbreviations

ASCO, American Society of Clinical Oncology; ASPS, alveolar soft part sarcoma; CI, confidence interval; CSF1, colony-stimulating factor 1; CSF1R, colony-stimulating factor 1 receptor; FDA, US Food and Drug Administration; GIST, gastrointestinal stromal tumor; HR, hazard ratio; NTRK, neurotrophic tyrosine receptor kinase; ORR, overall response rate; OS, overall survival; PVNS, pigmented villonodular synovitis; PDGFRA, platelet-derived growth factor receptor α; PFS, progression-free survival; STS, soft tissue sarcoma; TGCT, tenosynovial giant cell tumor; TKI, tyrosine kinase inhibitor; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor