GSJ Journal Club
Essential Reading From the Sarcoma and GIST Peer-Reviewed Literature

The peer-reviewed articles summarized in this section were selected by the Editorial Board for their timeliness, importance, relevance, and potential impact on clinical practice or translational research. 

Use of liquid biopsies to monitor disease progression in a sarcoma patient: a case report. Namlos HM, Zaikova O, Bierkehagen B, et al. BMC Cancer. 2017 Jan 6;17(1):29. doi: 10.1186/s12885-016-2992-8.Summary: Recent technological advances make it possible to use blood plasma containing circulating cell-free tumor DNA (ctDNA) as a liquid biopsy. In this case report we show how serial liquid biopsies can be used to monitor the disease course and detect disease recurrence in a sarcoma patient. A 55-year-old male presented with a rapidly growing, painful palpable mass in the left groin region, and a biopsy revealed a high-grade malignant spindle cell sarcoma. No metastases were detected on radiologic imaging scans. Using targeted resequencing with a custom 900 cancer gene panel, eight somatic mutations among them KRAS and NF1, were identified in the primary tumor. Targeted resequencing of plasma cell-free DNA (ctDNA) collected before and after surgery and at disease progression confirmed the presence of six of eight mutations at all three time points. The ctDNA level, estimated from the somatic allele frequencies of these six mutations, was high in plasma taken at the time of surgery, at levels similar to the primary tumor. Detection of low levels of ctDNA three days after surgery indicated persistent microscopic disease. Repeated radiologic imaging six weeks postoperatively showed widespread metastatic disease in the lungs, skeleton and the pelvic region. At this time point there was a dramatic increase in the ctDNA level, reflecting the disease progression of the patient. The patient had an unusually aggressive cancer, and succumbed to the disease 13 weeks after surgery.
Conclusion: This case report demonstrated that targeted resequencing of ctDNA from longitudinal collected plasma can be used to monitor disease progression in a soft tissue sarcoma patient, including manifestation of metastatic disease. The ctDNA represented the genomic profile of the tumor, supporting clinical use of liquid biopsies to identify tumor-specific mutations as well as recurrent disease.

Response to anti-PD1 therapy with nivolumab in metastatic sarcomas. Paoluzzi L, Cacavio A, Ghesani M, et al. Clin Sarcoma Res. 2016;Dec. 30:66-24.
Summary: Manipulation of immune checkpoints such as CTLA4 or PD-1 with targeted antibodies has recently emerged as an effective anticancer strategy in multiple malignancies. Sarcomas are a heterogeneous group of diseases in need of more effective treatments. Different subtypes of soft tissue and bone sarcomas have been shown to express PD-1 ligand. This study retrospectively analyzed a cohort of patients (pts) with relapsed metastatic/unresectable sarcomas, who were treated with nivolumab provided under a patient assistance program from the manufacturer. Pts underwent CT or PET/CT imaging at baseline and after at least four doses of nivolumab; RECIST 1.1 criteria were used for response assessment. Twenty-eight pts with soft tissue (STS, N = 24) or bone sarcoma (N = 4), received IV nivolumab 3 mg/kg every 2 weeks from July 2015. Median age was 57 (24-78), male:female ratio was 14:14; the median number of nivolumab cycles was eight. Eighteen pts concomitantly received pazopanib at 400-800 mg daily. The most common side effect was grade 1-2 LFT elevations; grade 3-4 toxicity occurred in five patients (colitis, LFT elevations, pneumonitis). Twenty-four pts received at least four cycles. We observed three partial responses: one dedifferentiated chondrosarcoma, one epithelioid sarcoma and one maxillary osteosarcoma (last two patients on pazopanib); nine patients had stable disease including three leiomyosarcomas; 12 patients had progression of disease including 4 leiomyosarcoma. Clinical benefit (response + stability) was observed in 50% of the evaluable patients.
Conclusion: These data provide a rationale for further exploring the efficacy of nivolumab and other checkpoint inhibitors in soft tissue and bone sarcoma.

Trabectedin for soft tissue sarcoma: current status and future perspectives. Gordon EM, Sanhala KK, Chawla N, et al. Adv Ther. 2016;33:1055-10771.
Summary: Trabectedin, derived from the marine ascidian, Ecteinascidia turbinata, is a natural alkaloid with multiple complex mechanisms of action. On 23 October 2015, 15 years after the results of the first Phase 1 clinical trial using trabectedin for chemotherapy-resistant solid malignancies was reported, and 8 years after its approval in Europe, the United States Food and Drug Administration (USFDA) finally approved trabectedin for the treatment of unresectable or metastatic liposarcoma or leiomyosarcoma that has failed a prior anthracycline-containing regimen. Approval was based on the results of a pivotal Phase 3 trial involving a 2:1 randomization of 518 patients (who were further stratified by soft tissue sarcoma subtype), in which a significant improvement in progression-free survival was reported in the trabectedin-treated group vs. the dacarbazine-treated group (P < 0.001). In this trial, the most common adverse reactions were nausea, fatigue, vomiting, constipation, anorexia, diarrhea, peripheral edema, dyspnea, and headache, while the most serious were neutropenic sepsis, rhabdomyolysis, cardiomyopathy, hepatotoxicity, and extravasation leading to tissue necrosis. The most common grade 3-4 adverse events were laboratory abnormalities of myelosuppression in both arms and transient transaminitis in the trabectedin arm.
Conclusion: In a recent Phase 2 trial, trabectedin had a similar outcome as doxorubicin when given as a single agent in the first-line setting. Studies are also being conducted to expand the use of trabectedin not only as a first-line cancer drug, but also for a number of other clinical indications, for example, in the case of mesenchymal chondrosarcoma, for which trabectedin has been reported to be exceptionally active. The possibility of combining trabectedin with targeted therapies, immune checkpoint inhibitors or virotherapy would also be an interesting concept. In short, trabectedin is an old new drug with proven potential to impact the lives of patients with soft tissue sarcoma and other solid malignancies.

Olaratumab and doxorubicin versus doxorubicin alone for treatment of soft-tissue sarcoma: an open-label phase 1b and randomised phase 2 trial. Tap WD, Jones RL, Van Tine BA, et al. Lancet. 2016;388:488-497.
Summary: Treatment with doxorubicin is a present standard of care for patients with metastatic soft-tissue sarcoma and median overall survival for those treated is 12-16 months, but few, if any, novel treatments or chemotherapy combinations have been able to improve these poor outcomes. Olaratumab is a human antiplatelet-derived growth factor receptor α monoclonal antibody that has antitumor activity in human sarcoma xenografts. The study assessed the efficacy of olaratumab plus doxorubicin in patients with advanced or metastatic soft-tissue sarcoma. We did an open-label phase 1b and randomised phase 2 study of doxorubicin plus olaratumab treatment in patients with unresectable or metastatic soft-tissue sarcoma at 16 clinical sites in the USA. For both the phase 1b and phase 2 parts of the study, eligible patients were aged 18 years or older and had a histologically confirmed diagnosis of locally advanced or metastatic soft-tissue sarcoma not previously treated with an anthracycline, an Eastern Cooperative Oncology Group (ECOG) performance status of 0-2, and available tumor tissue to determine PDGFRα expression by immunohistochemistry. In the phase 2 part of the study, patients were randomly assigned in a 1:1 ratio to receive either olaratumab (15 mg/kg) intravenously on day 1 and day 8 plus doxorubicin (75 mg/m(2)) or doxorubicin alone (75 mg/m(2)) on day 1 of each 21-day cycle for up to eight cycles. Randomization was dynamic and used the minimization randomization technique. The phase 1b primary endpoint was safety and the phase 2 primary endpoint was progression-free survival using a two-sided level of 0.2 and statistical power of 0.8. This study was registered with, number NCT01185964.
Findings: 15 patients were enrolled and treated with olaratumab plus doxorubicin in the phase 1b study, and 133 patients were randomised (66 to olaratumab plus doxorubicin; 67 to doxorubicin alone) in the phase 2 trial, 129 (97%) of whom received at least one dose of study treatment (64 received olaratumab plus doxorubicin, 65 received doxorubicin). Median progression-free survival in phase 2 was 6.6 months (95% CI 4.1-8.3) with olaratumab plus doxorubicin and 4.1 months (2.8-5.4) with doxorubicin (stratified hazard ratio [HR] 0.67; 0.44-1.02, P=0.0615). Median overall survival was 26.5 months (20.9-31.7) with olaratumab plus doxorubicin and 14.7 months (9.2-17.1) with doxorubicin (stratified HR 0.46, 0.30-0.71, P=0.0003). The objective response rate was 18.2% (9.8-29.6) with olaratumab plus doxorubicin and 11.9% (5.3-22.2) with doxorubicin (p=0.3421). Steady state olaratumab serum concentrations were reached during cycle 3 with mean maximum and trough concentrations ranging from 419 mg/mL (geometric coefficient of variation in percentage [CV%] 26.2) to 487 μg/mL (CV% 33.0) and from 123 μg/mL (CV% 31.2) to 156 μg/mL (CV% 38.0), respectively. Adverse events that were more frequent with olaratumab plus doxorubicin versus doxorubicin alone included neutropenia (37 [58%] vs 23 [35%]), mucositis (34 [53%] vs 23 [35%]), nausea (47 [73%] vs 34 [52%]), vomiting (29 [45%] vs 12 [18%]), and diarrhea (22 [34%] vs 15 [23%]). Febrile neutropenia of grade 3 or higher was similar in both groups (olaratumab plus doxorubicin: eight [13%] of 64 patients vs doxorubicin: nine [14%] of 65 patients).
Conclusion: This study of olaratumab with doxorubicin in patients with advanced soft-tissue sarcoma met its predefined primary endpoint for progression-free survival and achieved a highly significant improvement of 11.8 months in median overall survival, suggesting a potential shift in the treatment of soft-tissue sarcoma.

Clinical implications of repeated drug monitoring of imatinib in patients with metastatic gastrointestinal stromal tumour. Hompland I, Bruland OS, Ubhayasek-hera, et al. Clin Sarcoma Res. 2016;21;Dec. DOI:10 1186/s13569-016-0062-2.
Summary: Imatinib mesylate (IM) is the preferred treatment for the majority of patients with metastatic gastrointestinal stromal tumour (GIST). Low trough IM concentration (Cmin) values have been associated with poor clinical outcomes in GIST patients. However, there are few studies of repeated measurements of IM levels, and therapeutic drug monitoring is not yet a part of routine clinical practice. This study was conducted to reveal clinical scenarios where plasma concentration measurement of IM trough level (Cmin) is advantageous. Patients with advanced GIST receiving IM were included from January 2011 to April 2015. Heparin plasma was collected at each follow-up visit. Ninety-six samples from 24 patients were selected for IM concentration measurement. Associations between IM plasma concentration and clinical variables were analyzed by Students’ t test, univariate and multivariate linear regression analyses. The mean IM Cmin plasma concentrations for patients taking <400, 400 and >400 mg daily were 782, 1132 and 1665 ng/mL, respectively (P = 0.010). High IM Cmin levels were correlated with age, low body surface area, low hemoglobin concentration, low creatinine clearance, absence of liver metastasis and no prior gastric resection in univariate analysis. In multivariate analysis age, gastric resection and liver metastasis were included in the final model. Eight patients had disease progression during the study, and mean IM levels were significantly lower at time of progression compared to the previous measurement for the same patients (770 and 1223 ng/mL, respectively; P = 0.020).
Conclusion: The results do not support repeated monitoring of IM levels on a routine basis in all patients. However, results have revealed clinical scenarios where drug measurement could be beneficial, such as for patients who have undergone gastric resection, suspicion of non-compliance, subjectively reported side effects, in elderly patients and at the time of disease progression.

High nuclear expression of proteasome activator complex subunit 1 predicts poor survival in soft tissue leiomyosarcomas. Lou S, Cleven AHG, Baluff B, et al. Clin Sarcoma Res. 2016;6:17
Summary: Previous studies on high grade sarcomas using mass spectrometry imaging showed proteasome activator complex subunit 1 (PSME1) to be associated with poor survival in soft tissue sarcoma patients. PSME1 is involved in immunoproteasome assembly for generating tumor antigens presented by MHC class I molecules. In this study, we aimed to validate PSME1 as a prognostic biomarker in an independent and larger series of soft tissue sarcomas by immunohistochemistry. Tissue microarrays containing leiomyosarcomas (n = 34), myxofibrosarcomas (n = 14), undifferentiated pleomorphic sarcomas (n = 15), undifferentiated spindle cell sarcomas (n = 4), pleomorphic liposarcomas (n = 4), pleomorphic rhabdomyosarcomas (n = 2), and uterine leiomyomas (n = 7) were analyzed for protein expression of PSME1 using immunohistochemistry. Survival times were compared between high and low expression groups using Kaplan–Meier analysis. Cox regression models as multivariate analysis were performed to evaluate whether the associations were independent of other important clinical covariates. PSME1 expression was variable among soft tissue sarcomas. In leiomyosarcomas, high expression was associated with overall poor survival (P = 0.034), decreased metastasis-free survival (P = 0.002) and lower event-free survival (P = 0.007). Using multivariate analysis, the association between PSME1 expression and metastasis-free survival was still significant (P = 0.025) and independent of the histological grade.
Conclusion: High expression of PSME1 is associated with poor metastasis-free survival in soft tissue leiomyosarcoma patients, and might be used as an independent prognostic biomarker.  GSJ




The Fifth Pillar of Cancer Treatment: Is It Time to Explore Immunotherapy for GIST?

Breelyn A. Wilky, MD, MS

Breelyn A. Wilky, MD
Assistant Professor
Sylvester Comprehensive Cancer Center
at the University of Miami
Miller School of Medicine
Miami, Florida


Surgery, chemotherapy, and radiation have long been dubbed the three pillars of effective cancer therapy.  However, gastrointestinal stromal tumors (GIST) pose a unique challenge compared to other cancers of the gastrointestinal tract – they are resistant to traditional chemotherapy, and radiation treatment does not improve the risk of relapse after surgical resection.  Instead, GIST is one of the best examples of a cancer that benefits from the fourth pillar of cancer therapy – using targeted small molecules that block specific signaling pathways in cancer cells.

The hallmark of GIST is that interstitial cells of Cajal acquire pathologic activating mutations in the c-KIT receptor tyrosine kinase, leading to constitutive activation of the pathway. Accordingly, blockade of KIT signaling with tyrosine kinase inhibitors such as imatinib leads to clinical responses in nearly 80% of patients with GIST. However, 50% of patients with advanced GIST acquire resistance to imatinib within two years. Although other tyrosine kinase inhibitors like sunitinib, regorafenib, and others have demonstrated benefit in the imatinib-resistant setting, most patients eventually become resistant to tyrosine kinase inhibitors and are in desperate need of new therapies.

The reason for failure of most tyrosine kinase inhibitors is that GIST cells acquire secondary mutations that allow them to circumvent the blockade in KIT signaling, often utilizing other cellular pathways. The rapid development of resistance in cancer cells to either chemo-therapy or targeted therapy is one of the main reasons that immune-directed therapies are so appealing. At least in theory, the immune system has the adaptability to recognize and attack numerous different antigens, maintaining memory and constant surveillance. However, cancer cells have evolved various mechanisms to evade the immune system, avoiding detection and attack.

The focus of modern immunotherapy is to counteract these evasion mechanisms, and allow the immune system to recognize and attack cancer cells.The development of modern immunotherapy strategies, including checkpoint inhibitors and adoptive T cell therapies has led to remarkable responses in dozens of refractory solid and hematologic malignancies. With reports of long-term remissions in previously incurable diseases like metastatic melanoma, non-small cell lung cancer, and acute lymphoblastic leukemia, it is no surprise that immunotherapy is now being heralded as the fifth pillar of cancer care.

In this review, we will summarize the available laboratory data on the role of the immune system in GIST and highlight ongoing clinical trials focused on immune-directed treatments for GIST patients.

Overview of the Immune System and Potential
Applications of Immunotherap
The immune system is constantly autoregulating between pro-inflammatory and anti-inflammatory states, both globally and focally within tumor deposits. As summarized in the Figure, an effective anti-tumor response requires the following: 1. recognition of tumor-specific proteins called neo-antigens, 2. induction of the body’s innate and adaptive immune machinery, 3. effective migration of the key immune cells into the tumor deposits, and 4. maintenance of the attack at the tumor-immune cell interface.

Figure. An effective anti-tumor response requires the following: recognition of tumor-specific proteins called neoantigens, induction of the body’s innate and adaptive immunemachinery, effective migration of the key immune cells into the tumor deposits, and maintenance of the attack at the tumor-immune cell interface. Used with permission from Chen & Mellman, Immunity, 2013. To view a larger version of this figure, click here.

At every phase of immune activity, there are a variety of natural mechanisms to regulate the inflammatory response. This occurs through changes in pro- and anti-inflammatory cytokines, expression of suppressive checkpoint proteins on antigen-presenting cells and effector T cells, and emergence of suppressive phenotypes including exhausted or anergic T cells, T regulatory cells (Tregs), M2 macrophages, and myeloid-derived suppressor cells. This highly complex interplay has been reviewed in depth1,2 and is summarized in the Table.

To view a larger version of this table, click here.

Most cancer cells are initially recognized by the immune system, specifically cytotoxic CD8+ T cells, as foreign based on abnormal protein production from mutations in DNA, that lead to neoantigens. Because of this, most cancerous cells are initially immunogenic, meaning that they are recognized and destroyed by the immune system before forming clinically detectable tumors. However, some tumor cells may inherently be poorly recognized by the immune system, or they develop ways to avoid the immune system attack, known as immune evasion. Over time, the immunogenicity of the cancer cells changes, a phenomenon known as immunoediting.3 Ultimately, the less-immunogenic cancer cells will overcome the immune system leading to clinically detectable and immunoresistant tumors.

Cancer cells can evade immune system recognition and attack by three main mechanisms. First, immunoedited cancer cells tend to lose expression of key immunogenic neoantigens. They may also stop expressing the Class I MHC complex, which is required for recognition of neoantigens by cytotoxic CD8+ T cells. Secondly, cancer cells can produce the same immunosuppressive cytokines and express checkpoint proteins that normally function to blunt the immune response. For example, many tumors including GIST secrete vascular endothelial growth factor (VEGF) that not only facilitates angiogenesis, but also promotes an immunosuppressive tumor microenvironment that limits immune cell recruitment, infiltration and activation.

Finally, even if immune cells successfully infiltrate the tumors, cytotoxic T cells and macrophages may express check-point proteins that shut down further activation, evolve to an anergic state, or become suppressive phenotypes like Tregs. Most tumor cells directly promote an environment that is suppressive rather than inflammatory.

The focus of modern immunotherapy aims to counteract many of these evasion mechanisms, such as checkpoint inhibitors which block the immunosuppression resulting from the PD-1/PD-L1 interaction. However, checkpoint inhibitors are not effective for everyone, and counterintuitively, activity is not dependent on expression of PD-1/PD-L1 by the tumor cells. A better understanding of the mechanisms used to avoid immune recognition by a particular patient’s GIST may help us to customize immunotherapy approaches, similarly to how we use mutation testing to customize targeted therapy.

The Role of the Immune System in GIST –
Preclinical Evidence

Immune cells have the ability to infiltrate GIST tumor deposits, but lead to a generally suppressive environment.

One of the features that has been suggested to predict better prognosis, as well as a better response to immune-directed therapies in other types of cancer is the presence of infiltrating immune cells in metastatic tumor deposits.5  This information has led to the classification of tumors into “hot” and “cold” tumors, which likely require different types of immune-based therapies to induce immune responses.6  For example, a tumor without immune cell infiltration might be postulated to benefit from anti-VEGF therapies to try to optimize immune cell infiltration prior to induction of the immune activity with checkpoint inhibitors.

Our group has recently reviewed the reported literature evaluating immune cell infiltration in GIST.7,8  The most abundant infiltrating immune cells in GIST are tumor-associated macrophages (TAM), including alternatively-activated type 2 TAM and an immature macrophage phenotype expressing Ki-M1P.  CD3+ T cells and suppressive Tregs are also found in GIST, with a higher CD3/CD8 to Treg ratio correlating with improved outcomes.9 Interestingly, while the presence of immune cells tends to be associated with metastatic sites, unfavorable locations, and more aggressive proliferative in- dices, these tumors also tend to have superior progression-free survival. While overall the immune milieu seems to consist of largely suppressive phenotypes, it suggests that these tumors were at one point impacted by the immune system, and one could argue that immune therapies to reverse immunosuppression and restore the balance to a pro-inflammatory state may be more likely to impact the out- comes of these patients.

Imatinib Directly Impacts Immune Cells
in GIST Through Effects on IDO
The work from Dr Ronald DeMatteo and colleagues was recently presented at the American Society of Clinical Oncology (ASCO) 2016 Annual Meeting.  Using a mouse model of GIST, the animals were treated with imatinib and a variety of changes in the immune cells within the tumors were noted.9  They found that in tumors treated with imatinib, a much higher ratio of anti-tumor CD8+ cytotoxic T cells to suppressive Tregs was seen.  After looking at changes in mRNA in these treated tumors, a marked reduction in transcription of the enzyme indoleamine 2,3-dioxygenase (IDO), which catalyzes the conversion of tryptophan into metabolites that stabilize Tregs and suppress cytotoxic effector T cell activity was observed. Expression of IDO1 and a down-stream protein, KYN, were reported to be associated with improved responses to chemotherapy in soft tissue sarcomas at ASCO 2016 by Dr Toulmonde and colleagues (Abstract 11008).

Dr DeMatteo’s team went on to treat mice with MT-1, a more specific inhibitor of IDO. While IDO inhibitors alone produced only modest slowing of tumor growth, these effects were dramatically increased when combined with checkpoint inhibitors. Based on this observation, mice were treated with combination therapy with imatinib plus a checkpoint inhibitor of CTLA-4. Mice demonstrated less tumor growth when they received the combination of imatinib plus CTLA-4 inhibitor compared to either drug alone. This concept was recently explored in a clinical trial of dasatinib plus ipilimumab for GIST patients.

The results of this phase I study were recently reported, and showed 7/13 patients having partial responses by Choi criteria and an additional 3/13 patients with stable disease.10 Interestingly, one patient with PDGFR D842V mutation has achieved stable disease and remains on study for greater than 13.9 months.  Another patient with SDH-deficient GIST achieved Choi partial response and remained on study for 47 weeks. However, most patients had previously received multiple TKIs and were resistant to KIT inhibitors which likely lessened the potential effects of the combination therapy.  Further investigations of alternative KIT inhibitors, or studies with anti-PD-1 checkpoint inhibitors are ongoing or in planning stages.

Expression of Checkpoint Proteins in GIST Tumors
The expression of checkpoint proteins by human GIST cells including PD-L1 and TIM-3 has now been reported in several articles and reviewed in detail elsewhere.7,8,11 Dr. DeMatteo and colleagues also evaluated the presence of check- point proteins on tumor-infiltrating immune cells in human GIST samples collected at surgical excision.12 A higher proportion of tumor-infiltrating lymphocytes expressed the checkpoints PD-1, LAG-3, and TIM-3 relative to circulating immune cells. Higher expression of checkpoint proteins was seen in tumors that were resistant to imatinib.

A striking feature was that patients were highly variable in the degree of expression; some patients had very high levels of PD-1 whereas others had very low expression.  As seen in other cancers, this suggests that there may be particular patients with very high expression of checkpoint proteins along with immune cell infiltration, and it is tempting to speculate that these patients would be most critical to include in clinical trials exploring checkpoint inhibitors and other types of immunotherapy.

It is still unclear whether checkpoint expression on tumor cells as compared to expression on infiltrating immune cells is more important for effect from checkpoint inhibitors.  In many cancers, expression of PD-L1 or PD-1 on immune effector cells predicts response to PD-1 antibodies, including bladder cancer and non-small cell lung cancer, however in melanoma, patients lacking expression can still respond to these drugs.  We are awaiting the follow-up correlative data from the recently reported Phase II clinical trial, SARC028, studying pembrolizumab in patients with bone and soft tissue sarcomas.  These results will determine whether responders in this trial exhibited tumor or TIL PD-1/PD-L1 expression. GIST patients were not included in this study, thus biomarkers of GIST patients who may respond to immunotherapy are still unknown.

VEGF May Impact the Immune System in GIST
Vascular endothelial growth factor (VEGF) and related cofactors are known to drive tumor angiogenesis, and the resulting aberrant blood vessel networks are critical for tumor growth and metastasis. GIST also depends on VEGF signaling. High expression of VEGF correlates with inferior PFS after treatment with imatinib,13 and the most effective therapies in imatinib-resistant GIST are TKIs that also inhibit VEGFR, like sunitinib and regorafenib. Critically, Dmitry Gabrilovich and others have shown that VEGF and related proteins like HIF-1αcan directly suppress immune responses within the tumor microenvironment by inhibiting dendritic cell maturation and antigen presentation, blocking migration of lymphocytes across endothelium into tumor deposits, and promoting accumulation of suppressive myeloid-derived suppressor cells (MDSC), tumor-associated macro-phages (TAM) and Tregs.14-18

As we have discussed previously, GIST that is resistant to imatinib shows upregulation of checkpoint proteins compared to untreated tumors. Thus, one may speculate that the worse survival of GIST patients with high VEGF may be related to suppression of the immune system. By inhibiting VEGF, we might improve immune cell infiltration into the tumors which has been shown to be a positive factor in GIST outcomes. Combinations of VEGF blockade plus immune checkpoint inhibitors has improved anti-tumor response in other cancers like melanoma and renal cell carcinoma.19,20  This is the rationale for our ongoing Phase II clinical trial of the VEGFR inhibitor axitinib plus pembrolizumab, which is currently open to accrual and includes GIST patients (NCT02636725).

Adoptive T cell Therapy for GIST
Another strategy for immunotherapy is the concept of adoptive T cell therapy. Rather than relying on the native immune system’s ability to recognize antigens of interest and to stimulate an effective adaptive response, adoptive T cell therapy bypasses these steps by engineering a patient’s own T cells to express a T cell receptor specific for the target of interest. The expanded T cell population is then returned to the patient. This strategy has been used effectively in acute lymphoblastic leukemia and lymphomas and has been shown to produce patient responses in NY-ESO-1 positive synovial sarcomas.21

Dr Steven Katz and colleagues have recently reported the development of a KIT-specific chimeric antigen receptor (CAR) for transduction into human T cells.22 These engineered T-cells were effective at suppressing GIST growth in mice models. While these advances are promising, the reported toxicities with adoptive T cell therapy from cross-reaction against normal human tissues are not trivial, and further testing and development will be required before these strategies are ready for clinical testing.

Clinical Opportunities for the Exploration
of Immunotherapy in GISTs
With the excitement surrounding immunotherapy in medical conferences as well as the media, a key question is when GIST patients and their providers might want to consider participating in a clinical trial. This decision is particularly difficult given that many trials require patients to stop using tyrosine kinase inhibitors which we generally continue without interruption even in the setting of resistant disease. GIST patients in general do best with uninterrupted suppression of KIT, even when the disease has developed resistance pathways.  We have learned that response to checkpoint inhibitors can take months to manifest, and patients with GIST who abruptly discontinue TKIs can often experience a flare and rapid growth of their disease. Even if patients ultimately respond to immunotherapy, the short-term interval growth can be problematic and lead to medical complications like gastrointestinal obstruction. While laboratory evidence is promising for potential benefit from immunotherapy for GIST, there is still minimal clinical data to guide patients and physicians in whether the science will translate to patient responses in the clinic.

Additionally, a common misconception is that since immunotherapy is designed to boost the patient’s own natural immune system to fight cancer, the side effects of these treatments will be less toxic than chemotherapy or targeted therapies. While most patients on checkpoint inhibitors tolerate these drugs well, there can be significant auto-immune toxicities that can lead to life-threatening inflammation or even death, particularly with checkpoint inhibitor combinations. Adoptive T cell therapy can be even more toxic, and the ongoing trials of CAR T cells and other engineered T cell therapies have led to treatment-related deaths in some patients from overwhelming inflammation. It is critical that patients and providers are educated on the potential adverse reactions prior to embarking on clinical trials studying these agents.

Regardless, for GIST patients without other treatment options, there are now several clinical trials that utilize immunotherapy, especially novel combinations involving tyrosine kinase inhibitors, that will help to answer whether immunotherapy will have a role in management of advanced GIST. The list below shows clinical trials for which GIST patients are eligible. Additional information can be found at

  • Phase I, imatinib plus ipilimumab (NCT01738139). Offered at MD Anderson Cancer Center. Metastatic or unresectable GIST, patients with mutations in exon 13 V654X, 14 T6701, 17 D816X and all exon 18 muta-tions will not be eligible for enrollment.
  • Phase II (randomized), nivolumab (anti-PD-1) with or without ipilimumab (anti-CTLA4) for patients with metastatic GIST, (NCT02880020). Offered at UCLA.  GIST that is unresectable and has progressed on imatinib are eligible.
  • Phase II, metronomic cyclophosphamide and pembrolizumab (PEMBROSARC), (NCT02406781). Offered in various locations in France. GIST-specific arm, progression after imatinib and sunitinib.
  • Phase II, durvalumab (anti-PD-L1) and tremeli-mumab (anti-CTLA4), (NCT02815995). Offered at MD Anderson Cancer Center. GIST eligible for “other sarcoma” arm.
  • Phase I/IIa, pexidartinib (PLX3397, inhibits KIT among other targets) plus pembrolizumab (NCT024 52424). Offered in Arizona, California, Michigan, Texas.  GIST eligible for Phase II portion only.
  • Phase II, axitinib (anti-VEGFR) plus pem-brolizumab (NCT02636725). Offered at Sylvester Comprehensive Cancer Center, Miami. Metastatic/unresectable GIST refractory to at least one TKI are eligible.
  • Phase I, Intuvax (intratumor injections) plus suni-tinib, NCT02686944. Offered in Sweden. Metastatic/unresectable GIST refractory to imatinib and sunitinib.
  • Phase I, enoblituzumab (anti-B7-H3) and pembro-lizumab, (NCT02475213). All solid tumors eligible but must be B7-H3 positive (requires additional testing.) Offered in Texas, Maryland, New York, Pennsylvania, and Florida.
  • Phase I, TLR4 Agonist GLA-SE and Radiation Therapy (NCT02180698). Offered at Fred Hutchinson Cancer Center (Seattle). Open for various types of soft tissue sarcoma.
  • Phase I, GDC0919 (IDO inhibitor), (NCT02048709). Offered in Georgia. All solid tumors eligible.
  • Phase I/II, In Situ vaccination with tremelimumab (anti-CTLA4) and IV durvalumab (anti-PD-L1) plus PolyICLC, (NCT02643303). Offered at Ludwig Institute, New York. All solid tumors.

Summary and Conclusions
In summary, modern immunotherapy has generated incredible excitement and hope for patients with cancer that has not responded to traditional chemotherapy or targeted treatments. However, there is still much that is unknown about which patients are most likely to benefit. Gastrointestinal stromal tumors are among the best-studied sarcomas in the laboratory with significant evidence that the immune system is a critical component of the body’s ability to repress and attack GIST.

Thus, the investigation of immunotherapy as a potential treatment strategy to fight GIST is critical. The decision to participate in a clinical trial is not one that is to be taken lightly, particularly given the relative newness of modern immunotherapy drugs and the significant toxicities that can occur. However, with the increasing number of scientifically-driven immunotherapy clinical trials, GIST patients now have the opportunity to take part in one of the newest frontiers in cancer research.

1. Muenst S, Läubli H, Soysal SD, Zippelius A, Tzankov A, and Hoeller S. The immune system and cancer evasion strategies: therapeutic concepts. J Intern Med. 2016; 279:541-62.
2. Chen DS and Mellman I. Oncology meets immunology: the cancer-immunity cycle. Immunity. 2013; 39:1-10.
3. Schreiber RD, Old LJ, and Smyth MJ. Cancer immunoediting: integrating immunity’s roles in cancer suppression and promotion. Science. 2011; 331:1565-70.
4. Lee A, Huang P, DeMatteo RP, and Pollack SM. Immunotherapy for Soft Tissue Sarcoma: Tomorrow Is Only a Day Away. Am Soc Clin Oncol Educ Book. 2016; 35:281-90.
5. Jochems C and Schlom J. Tumor-infiltrating immune cells and prognosis: the potential link between conventional cancer therapy and immunity. Exp Biol Med. 2011; 236:567-579.
6. Gajewski TF, Schreiber H, and Fu Y-X. Innate and adaptive immune cells in the tumor microenvironment. Nat Immunol. 2013; 14:1014-1022.
7. Tan T, Trent JC, Wilky BA, Kerr DA, and Rosenberg AE. Current Status of Immunotherapy for Gastrointestinal Stromal Tumor. Cancer Gen Ther. 2016; In press.
8. Tan Y, Garcia-Buitragoa MT, Trent JC, and Rosenberg AE. The immune system and gastrointestinal stromal tumor: a wealth of opportunities. Curr Opin Oncol. 2015; 27:338-42.
9. Balachandran VP, Cavnar MJ, Zeng S, Bamboat ZM, Ocuin LM, Obaid H, et al. Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the inhibition of Ido. Nat Med. 2011; 17:1094-100.
10. D’Angelo SP, Shoushtari AN, Keohan ML, Dickson MA, Gounde MM, Chi P, et al. Combined KIT and CTLA-4 Blockade in Patients with Refractory GIST and Other Advanced Sarcomas: A Phase Ib Study of Dasatinib Plus Ipilimumab. Clin Cancer Res. 2016.
11. D’Angelo SP, Shoushtari AN, Agarm NP, Kuk D, Qin lL-X, Carvajal RD, et al. Prevalence of tumor-infiltrating lymphocytes and PD-L1 expression in the soft tissue sarcoma microenvironment. Hum Pathol. 2014; 46:357-65.
12. Seifert AM, Zeng S, Zhang JQ, Kim TS, N.A.Cohen, Beckman MJ, et al. PD-1/PD-L1 blockade enhances T cell activity and antitumor efficacy of imatinib in gastrointestinal stromal tumors. Clin Cancer Res. 2016.
13. McAuliffe JC, Lazar AJ, Yang D, Steinert DM, Qiao W, Thall PF, et al. Association of intratumoral vascular endothelial growth factor expression and clinical outcome for patients with gastrointestinal stromal tumors treated with imatinib mesylate. Clin Cancer Res. 2007; 13:6727-34.
14. Finke JH, Rini B, Ireland J, Rayman P, Richmond A, Golshayan A, et al. Sunitinib reverses type-1 immune suppression and decreases T-regulatory cells in renal cell carcinoma patients. Clin Cancer Res. 2008; 14:6674-82.
15. Gabrilovich DI, Chen HL, Girgis KR, Cunningham HT, Meny GM, Nadaf S, et al. Production of vascular endothelial growth factor by human tumors inhibits the functional maturation of dendritic cells. Nat Med. 1996; 2:1096-103.
16. Xin H, Zhang C, Herrmann A, Du Y, Figlin R, and Yu H. Sunitinib inhibition of Stat3 induces renal cell carcinoma tumor cell apoptosis and reduces immunosuppressive cells. Cancer Res. 2009; 69:2506-13.
17. Griffioen AW, Damen CA, Martinotti S, Blijham GH, and Groenewegen G. Endothelial intercellular adhesion molecule-1 expression is suppressed in human malignancies: the role of angiogenic factors. Cancer Res. 1996; 56:1111-17.
18. Kumar V and Gabrilovich DI. Hypoxia-inducible factors in regulation of immune responses in tumour microenvironment. Immunology. 2014; 143:512-519.
19. Amin A, Plimack ER, Infante JR, Ernstoff MS, Rini BI, McDermott DF, et al. Nivolumab (anti-PD-1; BMS-936558, ONO-4538) in combination with sunitinib or pazopanib in patients (pts) with metastatic renal cell carcinoma (mRCC). J Clin Oncol. 2014; 32:abstract 5010.
20. Hodi FS, Lawrence D, Lezcano C, Wu X, Zhou J, Sasada T, et al. Bevacizumab plus Ipilimumab in Patients with Metastatic Melanoma. Cancer Immunol Res. 2014; 2:632-42.
21. D’Angelo S, Merchant M, Melchiori L, Zhang H, Pandite L, Holdich T, et al. Optimizing engineered TCR T cell therapy for synovial sarcoma. J Immunother Cancer. 2015; 3:P159.
22. Katz SC, Burga RA, Naheed S, Licata LA, Thorn M, Osgood D, et al. Anti-KIT designer T cells for the treatment of gastrointestinal stromal tumor. J Transl Med. 2013;11.  GSJ

Key words: immunotherapy, GIST, aKIT signaling, checkpoint inhibitors, adoptive T cell therapies, PD-1/PD-L1, pembrolizumab, imatinib.
Address for correspondence: Breelyn A. Wilky, MD, Assistant Professor – Sarcoma Program, Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine,1120 NW 14th Street, Suite 610E, Miami, FL 33136 E-mail:



Highlights From a Year With Significant Advances in Sarcoma Therapy and Related Progress in Detecting Recurrences, Mutational Subtypes of GIST

A recap of new information presented in the GIST and Sarcoma Journal suggests trends to follow in ongoing and future clinical trials. The information in this section represents summaries of the full articles published during the year. The complete articles can be viewed on the GIST and Sarcoma website—

Imaging to Detect Recurrent GIST After Surgery:
Controversies, Consensus and Guidelines
New recently published information, much of it based on expert consensus opinion and guidelines, is removing much of the uncertainty surrounding the optimal approach to detect recurrent gastrointestinal stromal tumor (GIST).  Although patients with GIST are generally followed up after surgery with longitudinally repeated imaging examinations to maximize the earliest detection of recurrence, there have been relatively few definitive studies on followup, the optimal approach for doing so may be unknown to many clinicians, and there has been a wide variation in guidelines.

All of this is changing rapidly with new reports clarifying our options, according to Wilbur B. Bowne, MD, Associate Professor of Surgery at Drexel University, Philadelphia. Until recently, however, there was a gap in the literature regarding optimal strategies because prospective studies were not conducted to investigate different follow-up schedules and methods. Key evidence is emerging on the rationale from groups like the National Comprehensive Cancer Network and the European Society for Clinical Oncology; additional expert opinions are available from consensus reports from other groups who have studied surveillance in high risk groups.  All of this information is building on a solid framework consisting of well recognized prognostic factors to help guide decision making.

Numerous reports have delineated the factors involved. The most important prognostic factor for recurrence is tumor proliferation rate, often assessed by counting of the number of mitotic figures per 50 high power fields (HPFs) of the microscope or by providing the number of mitotic figures per 1 mm2 of tumor. There are also other factors that are often independently associated with a high risk of recurrence, including non-gastric location of GIST, large size and tumor rupture.

Population-Based Study Provides Insight into GIST
Epidemiology and Risk of Additional Cancers
Although we have not completely deciphered the epidemiology of gastrointestinal stromal tumor (GIST) and additional cancers, a new population based study uses modern data collection resources to sort out significant temporal and disease associations between GIST and other cancers. Expanding upon the publications from several single institution studies, Jason Sicklick, MD and colleagues broadened the data set and cast a wider net to build upon these important earlier studies. Dr Sicklick is Assistant Professor of Surgery, Division of Surgical Oncology, Moores UCSD Cancer Center, University of California, San Diego, UC San Diego Health System.

Until recently, we relied upon the results of several descriptive, single-institution case series to determine the likelihood that patients with sporadic GISTs develop synchronous or metachronous malignancies. These studies qualitatively characterized cancer associations, but the findings were unsatisfying in their variability. For example, in a review by Agaimy and colleagues that included 4,813 patients, the frequency of additional malignancies varied from 4.5% to 33% in patients with GIST. There have been other large studies that added to this body of literature. For example, single or multiple institution studies have dem-onstrated associations between GIST and desmoids, acute myeloid leukemia, and other gastrointestinal malignancies found incidentally. While a hereditary etiology has been attributed to approximately 5% of all GIST cases associated with multiple benign and malignant tumors, Sicklick et al attempted to quantify the frequency and temporal relationships of GIST to other cancer histologies in the remaining 95% of GIST patients without known hereditary GIST syndromes. These single-institution studies were helpful but they did not provide the broad-based characterizations.

In his Editorial in the journal, Cancer, Dr Constantine Stratakis, Scientific Director of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, wrote that it is a “different world out there for clinical oncology, one that changes every day because of the availability of large patient cohorts and the advances of modern genetics.” Population-based studies take advantage of the growing repository of data to make new associations, draw novel conclusions and possibly shift thinking about the histology and mechanics of diseases like GIST. In our recently published study, Sicklick et al utilized population-level data in the United States in order to define demographic, clinical characteristics, and temporal factors associated with increased probability of developing additional malignancies. From data culled from the Surveillance, Epidemiology, and End Results (SEER) database, the authors identified further evidence supporting the existence of nonrandom associations between GIST and other malignancies. This report builds on the foundation created by these studies with the hypothesis that greater insight could be gained into possible associations between GIST and other primary tumors with a national cancer database inquiry. Aside from the impact the findings may have on understanding GIST from a purely epidemiological perspective, the associations have important clinical implications for future cancer screening and treatment strategies.

Key Findings
The key findings emerging from this study include the following:

  • One-in-5.8 patients with GIST will develop additional malignancies before and after their diagnosis.
  • When compared to the United States population, people with GIST had a 44% increased prevalence of cancers occurring before a GIST diagnosis and a 66% higher incidence of developing cancers after diagnosis.
  • Patients with GIST are more likely to develop many cancers, including other sarcomas, non-Hodgkin’s lymphoma, carcinoid tumors, melanoma, colorectal, esophageal, pancreatic, hepatobiliary, non-small cell lung, prostate and renal cell cancers.
  • Non-Hispanic patients had a higher incidence of other cancers before a GIST diagnosis.
  • Patients with tumors smaller than 10 cm had a higher probability of a second cancer than patients with larger tumors.
  • Patients with tumors smaller than 2 cm had the greatest likelihood of developing additional malignancies, both before and after diagnosis.
  • Patients diagnosed with GIST may warrant consideration for additional screenings based on the other cancers that they are most susceptible to develop.

Hereditary GIST: A Growing Awareness of
Mutational Subtypes Is Redefining Recommendations
for Familial Screening, Surveillance
KIT and PDGFRA  generally represent the molecular hallmark of GIST. But the recent identification of other molecular characteristics beyond KIT and PDGFRA are driving a new view of the disease. Identification of germline mutations underscores the need for patients with these familial risk factors to undergo genetic counseling to determine appropriate followup and management. As new reports elucidate distinctions between subtypes of hereditary GIST, we are discovering how these mutations of GIST differ clinically, pathologically, and behaviorally from sporadic gastric tumors, according to Joshua D. Schiffman, MD, Professor,  Department of Pediatrics and Adjunct Professor, Department of Oncological Sciences, School of Medicine at the University of Utah, Salt Lake City.

Deeper insights into the biology of gastrointestinal stromal tumors (GISTs) are reshaping perceptions about GIST, its genetic risk factors and a diverse set of mutations and genotype features with clinical implications. Although GIST typically affects patients over the age of 40 years, recognition of its epidemiology in children and young adults has been increasingly recognized even though these younger groups account for only 1.4% of patients with the tumor.

  • Though the majority of GISTs appear to arise sporadically, a number of families with high frequencies of GISTs have been reported and germline mutations have been identified. The true frequency of all GIST diagnoses has been difficult to determine because the definition of GIST was derived in 1990 before it was molecularly characterized. One United States report from the Surveillance, Epidemiology, and End Results (SEER) database indicated that, from 1992 to 2000, the yearly incidence rate in the United States was 6.8 cases per million.
  • Most GISTs occurring in adults are driven by activating mutations in the KIT or PDGFRA genes. New findings on molecular classification, however, have dramatically changed the nomenclature for various GISTs and contributed to an improved understanding of hereditary and familial factors. For example, 85% of GISTs in children and 10% to 15% of GISTs in adults are negative for KIT and PDGFRA mutations and are commonly referred to as wild-type (WT) GIST. Because these malignant neoplasms are rare, efforts to delineate their natural history and determine their response to treatment have been difficult. This is particularly true with regard to the use of kinase inhibitor therapies: WT GIST generally does not respond to tyrosine kinase therapy known to be effective in non-WT GIST. From institutional series and case reports WT GIST has been characterized as primarily affecting young females, multifocal and primarily having a gastric location.

The Trabectedin Story: How a New Combination
Is Shifting the Paradigm of Treatment in Advanced
Uterine Leiomyosarcoma (uLMS)
Novel strategies are needed for the treatment of advanced uterine leiomyosarcoma. Until recently, this has largely been an elusive goal given the well-recognized and daunting limitations of conventional chemotherapeutic regimens. Robust evidence from recent studies, however, suggests how a new combination is changing perspectives on this disease, offering improved insights into the tumor microenvironment and delineating mechanisms of action, according to Marilyn Huang, MD, MS, Assistant Professor Division of Gynecologic Oncology, University of Miami/ Sylvester Comprehensive Cancer Center, Miami, Florida.

Moderate chemosensitivity and propensity for recurrence despite early stage portending an overall poor prognosis are perhaps the hallmarks to describe the challenges underlying treatment considerations in metastatic u-LMS. The cornerstone of treatment remains surgical resection. Following surgery, cytotoxic chemotherapy is often considered due to risk of recurrence. Chemotherapy is commonly doxorubicin or combination therapy with gemcitabine and docetaxel, although other chemotherapeutic agents such as dacarbazine and ifosamide have also been investigated. Multiple studies have demonstrated the challenges of treatment: estimated median survival remains 12 to 15 months for u-LMS.

Uterine sarcomas comprise approximately 3-7% of all uterine malignant neoplasms, of which u-LMS are the most common. Unfortunately, regardless of stage, u-LMS signifies a poor prognosis with few viable chemotherapy options. There are challenges in the diagnosis of u-LMS compared to other uterine cancers. In the case of endometrial cancer, the tumor originates from the endometrial lining, causing early bleeding in its development in >90% of women. Sampling of the endometrium yields malignant cells and an early diagnosis is typically achieved. However, endometrial sampling for early u-LMS may not be diagnostic and is often not performed by gynecologists in the setting of suspected uterine leiomyomas. Frequently, uLMS is not diagnosed until postoperatively at time of pathology review. Furthermore, u-LMS has a high propensity for both direct invasion as well as hematogenous spread and is relatively chemoinsensitive, making the disease difficult to treat. Optimal management of u-LMS is challenging and typically involves a multidisciplinary team whose approach generally depends on the disease spread (ie, localized vs disseminated disease).

Novel Combination—Olaratumab + Doxorubicin—
Suggests Potential Paradigm Shift in Treatment
of Soft-Tissue Sarcoma
The elusive goal of a therapy producing a pronounced survival benefit along with an acceptable safety profile in advanced soft-tissue sarcoma may be within reach. Not with chemotherapeutic approaches per se, but by combining a monoclonal antibody with doxorubicin. A new fully accrued trial is building on evidence pointing toward inhibition of a more clearly defined signaling pathway, based on a report by Gary K. Schwartz, MD, Chief, Division of Hematology and Oncology, Columbia University Medical Center, New York.

If a phase 3 clinical trial now underway can confirm or even improve upon the results achieved in earlier phases with a novel drug combination, a pronounced overall survival benefit may soon be available for patients with metastatic soft-tissue sarcoma. The development of novel systemic treatments in soft-tissue sarcomas is challenging; there are more than 70 different histological subtypes, and these subtypes show a heterogeneous genetic composition and clinical behavior. Within the subtypes there are also differences in overall survival as measured from the start of the first systemic treatment for advanced disease.

Given the little progress in improvement of median overall survival in patients with advanced soft-tissue sarcoma, the positive findings from an open-label phase 1b and randomized phase 2 trial suggest a potential shift in the paradigm for treatment. The study by Tap et al is believed to be the first randomized study to show increased survival for patients with soft-tissue sarcoma treated with a drug added to the standard-of-care doxorubicin treatment over single-agent doxorubicin. Among the key findings:

  • A combination of olaratumab plus doxorubicxin improved both progression-free survival (PFS) and overall survival compared with doxorubicin only.
  • The improvement of 11.8 months in median overall survival is highly significant, a difference of 80% compared with doxorubicin alone. The improvement in PFS was 2.5 months, a 61% advantage over doxorubicin alone.

Results from this investigation and, potentially data from the ongoing phase 3 trial now fully accrued, highlight the importance of a signaling pathway in the biology of soft-tissue sarcomas and how this pathway may be inhibited by the use of olaratumab, a monoclonal antibody. Platelet-derived growth factor (PDGF) and PDGF receptor (PDGFR) signaling plays a significant part in mesenchymal biology, including mesenchymal stem cell differentiation, growth, and angiogenesis.There is additional evidence from other reports on its importance: this pathway is also involved in cancer through aberrant cellular signaling. More specifically, it is believed to play a role in modulating the tumor or stromal microenvironment and facilitating metastases in several malignancies.

If these postulates are correct, then orlaratumab could have a significant impact on altering the course of soft-tissue sarcoma. Olaratumab is a recombinant human immunoglobulin G subclass (IgG1) monoclonal antibody that binds PDGFRα, blocking PDGF-AA, PDGF-BB and PDGF-CC binding and receptor activation.  Preclinical experience with olaratumab alone or in combination with doxorubicin—with results derived from human sarcoma xenograft models—suggested antitumor activity.  GSJ