Soft tissue sarcomas (STSs) are rare malignant tumours, which account for approximately 1% of adult cancers and include more than 50 histologic subtypes (1,2). Although several treatments are available, the prognosis of metastatic STSs is poor, with a median overall survival (OS) of 12 months (3). Undifferentiated pleomorphic sarcoma (UPS) is characterized by no identifiable line of differentiation and multiple cellular patternless forms. The prognosis of patients with UPS with recurrence and metastasis is poor (4,5).

Pazopanib is an oral tyrosine kinase inhibitor (TKI) with activity against vascular endothelial growth factor receptor 1 (VEGFR1), VEGFR2, VEGFR3, platelet-derived growth factor receptor A (PDGFRA), PDGFRB, and stem cell factor receptor (KIT). Pazopanib has been approved for the treatment of advanced STSs (6). The phase III PALETTE trial included patients with non-adipocytic STSs and progressive disease after standard chemotherapy. Pazopanib improved progression-free survival (PFS) by 3 months than placebo in this trial. However, OS did not improve (2). The genetic features of patients with STSs who responded to pazopanib are unclear. Herein, we report the case of a patient with advanced UPS characterized by coamplification of PDGFRA, VEGFR2, and KIT genes who experienced a lasting and complete response to pazopanib.

A 51-year-old man presented with abdominal pain and was admitted to Mitsui Memorial Hospital (Tokyo, Japan) in March 2019, 10 months before visiting our hospital, Osaka International Cancer Institute (Osaka, Japan). Computed tomography (CT) revealed an unexplained intra-abdominal haemorrhage close to the pancreatic head. Pain and haemorrhage spontaneously improved with rest. One month before admission, the patient experienced swelling and pain in the left thigh. Magnetic resonance imaging (MRI) revealed a 6.5-cm femoral bone tumour in the left thigh with haemorrhage and cortex osteolysis (Fig. 1A). Furthermore, 3 weeks before visiting our hospital, the abdominal pain reoccurred. CT revealed a retroperitoneal tumour with bleeding located in the groove adjacent to the pancreatic head (Fig. 1B). Hypermetabolism in the retroperitoneal tumour was detected using positron emission tomography-CT (PET/CT) at the time of the hospital visit. In addition, pancreatic head lymph nodes, second/fifth thoracic vertebrae, and left femoral bone tumours were detected (Fig. 1C and D). Blood test did not show any abnormal condition, such as elevated inflammatory response, tumour marker changes, or anaemia. Histological analysis of retroperitoneal and left thigh tumour biopsies revealed pleomorphic cell proliferation in a haphazard arrangement (Fig. 2). The tumour cells were focally positive for αSMA and CK AE1/AE3 and negative for desmin, h-caldesmon, S-100, SOX10, and MDM2. Based on the morphology, immune profile, and clinical presentation, the patient was diagnosed with UPS arising from the retroperitoneum with bone and lymph node metastases.

To prevent pathological fractures due to bone metastases, the patient was treated with denosumab and radiation therapy (35 Gy in 5 fractions) of the left femur and second/fifth thoracic vertebrae. The patient was then treated with 75 mg/m² doxorubicin administered on day 1 of a 21-day cycle for 8 cycles. Due to fatigue, nausea, and vomiting, the dose was reduced to 80% and then to 60%. After chemotherapy, the retroperitoneal primary lesion and bone metastases shrank and no new lesions emerged, suggesting tumour partial response (PR) (Fig. 3A). To determine the second-line chemotherapy after doxorubicin reached the upper limit, a comprehensive genomic profiling test (FoundationOne^® CDx, Foundation Medicine, Inc., Cambridge, MA, USA) was performed when the patient was receiving doxorubicin. FoundationOne CDx gene profiling can comprehensively identify 324 gene mutations using a next-generation sequencer and evaluate biomarkers, such as microsatellite status and tumour mutation burden (detailed information available at https://www.foundationmedicine.com/genomic-testing/foundation-one-cdx). Coamplification of the PDGFRA, VEGFR2, and KIT genes was detected (Table I). Pazopanib treatment was recommended by the expert, which is an inhibitor of receptor tyrosine kinases (RTKs), based on VEGFR2 amplification. Additionally, because pazopanib had inhibitory effects on VEGFR2, PDGFRA and KIT, it could be suitable. Therefore, 600 mg/day pazopanib was administered as second-line chemotherapy. However, the dose was reduced to 400 mg/day because of severe fatigue and diarrhoea. Surprisingly, the retroperitoneal primary tumour shrank more, and no new lesions developed for 3 years after the start of pazopanib treatment, indicating complete response (CR) (Fig. 3B and C). We performed follow-up imaging of the entire body by MRI and CT instead of PET-CT and did not observe any recurrence or metastasis after CR. Although grade 1 hypothyroidism and grade 2 hypertension were observed during 400 mg/day pazopanib treatment, no serious adverse events such as grade 3 or higher cytopenia occurred. Spontaneous pneumothorax is an evident and complicated side effect of pazopanib, and its occurrence is associated with lung metastasis (7,8). However, in this case, no lung metastasis or spontaneous pneumothorax was observed. Denosumab was discontinued within 3 years of treatment initiation.

Matching treatment strategies with tumour biology is a central principle of precision medicine, and the ectopic activation of RTKs is a universal theme in oncogenesis. Oncogenic fusions involving RTKs such as anaplastic lymphoma kinase (ALK) and proto-oncogene 1 (ROS1), and small deletions in epidermal growth factor receptor (EGFR) predict patient response to target-matched TKIs, especially in the context of non-small-cell lung cancer (9,10). Amplification of wild-type RTK genes, such as human epidermal growth factor receptor 2 (HER2), in breast and gastric cancers drives oncogenesis, and RTKs are therapeutic targets (11,12). Thus, the coamplification of PDGFRA, VEGFR2, and KIT, which is related to chromosome 4q12 amplification, may be an oncogenic driver and therapeutic target (13). Phase II trial of axitinib, a TKI of PDGFRs, VEGFR2, and KIT, included patients with recurrent adenoid cystic carcinoma. The longest responder (nearly four times longer than the median PFS for the study cohort) in this trial was a patient with amplified PDGFRA, VEGFR2, and KIT (14). We detected the coamplification of PDGFRA, VEGFR2, and KIT in UPS. Therefore, we treated the patient with pazopanib with second-line therapy. Interestingly, CR was achieved and maintained for 3 years with pazopanib. To the best of our knowledge, this is the first report to demonstrate that pazopanib was effective against STS with coamplified PDGFRA, VEGFR2, and KIT.

PDGFRA, VEGFR2, and KIT genes are located on chromosomal locus 4q12; hence, the concurrent amplification of these genes may be due to the overall amplification of chromosome 4q12. Coamplification of these RTKs is present in 0.86% of TCGA cases across all cancers and is more common in sarcomas and central nervous system neoplasms (13). In the Sarcoma Genome Project dataset, putative high-level amplifications of PDGFRA, VEGFR2, and KIT were reported in 3, 2 and 2.4%, respectively, of cancer cases (15). Although rare in absolute numbers, the coamplification frequency of these genes is consistent with other uncommon but highly targetable oncogenic kinase alterations, such as NTRK (16).

Pazopanib, which is the only approved noncytotoxic STS therapy, is an orally available inhibitor of multiple RTKs, including VEGFR1-3, PDGFRA/B, FGFR1/3/4, and KIT. The antitumor activity of pazopanib is attributed to antiangiogenic effects and inhibition of pro-proliferative signals mediated by RTKs on the surface of STS cells (17-19). Several clinical trials demonstrated survival benefits of pazopanib in patients with STS. In the PALETTE trial with 246 patients, none of the patients with STS had CR, 6% (14) had PR, 67% (164) had stable disease (SD), and 23% (57) had progressive disease (PD) (2). In the placebo arm, 0% (0 of 123) of patients had CR, 0% (0) had PR, 38% (47) had SD, and 57% (70) had PD (2). In a study by the Japanese Musculoskeletal Oncology Group, none of the 125 enrolled patients achieved CR (8). Therefore, achieving CR with pazopanib is rare in patients with STS. CR can be obtained in this case; however, it was not maintained in advanced UPS as per our experience. Furthermore, no standard treatment strategy has been established for treating sarcoma. Therefore, making a decision regarding the discontinuation of pazopanib was challenging. However, because CR persisted for a long time, discontinuation of pazopanib will be an important issue for future consideration.

The challenge of establishing reliable biomarkers for responsiveness to pazopanib has been addressed by several studies. Retrospective analyses of two EORTC studies and a real-world cohort of pazopanib-treated patients revealed that specific histologic types, including synovial sarcoma and desmoplastic small round cell tumour, and clinical parameters, including good performance status, low or intermediate tumour grade, and a normal haemoglobin levels, correlated with better outcomes (20,21). Several studies also explored the importance of molecular characteristics in selecting patients who may benefit from pazopanib. Heilig et al (22) analysed the molecular profiles and clinical outcomes of sarcoma patients treated with pazopanib and demonstrated that VEGFR2-high, NTRK3-high, and IGF1R-low mRNA levels were independently associated with PFS. In patients with advanced STS who exhibited short-term high-grade PR or long-term SD, Suehara et al (23) demonstrated that amplified GLI1 and elevated PDGFRB phosphorylation levels were linked to high antitumor activity of pazopanib. Using gene panel sequencing in 19 pazopanib-treated patients, Koehler et al (24) demonstrated that TP53 mutations correlated with better outcomes after pazopanib therapy. The present case suggests that STSs with amplified PDGFRA, VEGFR2, and KIT are rare; however, patients with these features may benefit from pazopanib therapy. A limitation of this study was the absence of immunohistological analysis, including markers such as PDGFRA, VEGFR2, KIT, VEGF, and PDGF. Therefore, further studies are required to confirm the benefits of pazopanib in patients with coamplification of PDGFRA, VEGFR2, and KIT, and to establish predictive markers for pazopanib sensitivity.