Role of Chemotherapy, VEGFR Inhibitors, and mTOR Inhibitors in Advanced Perivascular Epithelioid Cell Tumors (PEComas)

Perivascular epitheliod cell tumors (PEComas) are rare mesenchymal neoplasms characterized by epitheliod, sometimes pleomorphic cells with a perivascular distribution (1). Within the group of soft-tissue sarcomas, the family of PEComas has only relatively recently been recognized as a distinct subtype with specific morphologic and IHC characteristics (2). PEComas include angiomyolipoma (AML), clear cell sugar tumor of the lung, and lymphangioleiomyomatosis (LAM), which are often characterized by a benign clinical course. In contrast, other entities such as epithelioid AML and malignant PEComa can have an aggressive clinical course and develop distant metastases. The latter are typically large in size, with mitosis, necrosis, and nuclear pleomorphism (1), although standard pathologic criteria to discriminate a benign or malignant variant are not established (3).

The activity of mTOR inhibitors both in LAM and in malignant PEComas was reported (4–7). Indeed, these tumors harbor mutations and loss of heterozygosity of TSC2 gene or more rarely of TSC1 gene and can be associated with tuberous–sclerosis complex or be sporadic. TSC1 and TSC2 gene products contribute to a molecular complex, which negatively regulates the mTOR complex 1 (mTORC1). As a consequence of TSC1 or TSC2 alterations, the mTOR pathway is constitutively activated. This gave the rationale to use mTOR inhibitors (e.g., sirolimus) in the clinical setting. However, given the rarity of these tumors and the few available data, a precise estimation of response rate and median progression-free survival (PFS) is not available. In addition, the role of conventional chemotherapy or other biological agents (i.e., antiangiogenic agents) is not established as there are no published prospective clinical trials or sufficiently large retrospective case series. The aim of this multicenter, international study was to report on the activity of chemotherapy, antiangiogenic drugs, and mTOR inhibitors in patients with advanced/metastatic PEComa.

This was a retrospective, international study including patients treated at 5 European sarcoma reference centers (Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, IT; Royal Marsden Hospital NHS Foundation Trust, London GB, Centre Léon Bérard, Lyon, FR; Institut Gustave Roussy, Villejuif, FR; Hospital Universitario Virgen del Rocio, Sevilla, ES) and within the Italian Rare Cancer Network. We reviewed all consecutive cases of PEComa, malignant PEComa, or epitheliod AML treated between January 2000 and June 2018 with systemic therapy (including cytotoxic chemotherapy, VEGFR inhibitors, and mTOR inhibitors) for locally advanced and/or metastatic disease. All diagnoses were made according to the 2013 WHO classification of Tumours of Soft Tissue and Bone by expert pathologists. In Italy, all cases were centralized and reviewed in Treviso General Hospital, Treviso, or at Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan. Institutional approval was obtained at each participating center before study initiation. Anonymized data for patient demographics and disease diagnosis, staging, treatment, clinical course, and outcomes were recorded. RECIST 1.1 were used to assess response (8) Any radiological reduction in the sum of the longest diameter of target lesions that did not reach the criteria for an objective partial response (PR) was defined as a minor response. Authorization from reviewing ethics committees was obtained according to the local rules of each institution. The study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from patients alive at the time of data collection and analysis.

Survival analyses were performed using the Kaplan–Meier method and the Cox hazards regression models. PFS and overall survival (OS) were censored at the time of the last follow-up. Events for PFS were the date of the first documented evidence of progressive disease (PD) from each treatment start. OS event was considered to be death from any cause. Statistical analyses were performed using Prisma software and R software (v3.5.0). When corresponding P values were below the 5% threshold, statistical tests were considered significant.

Overall, 53 patients with a diagnosis of PEComa treated with systemic therapy for locally advanced/metastatic disease were identified. Patients and disease characteristics are summarized in Table 1. Median age was 54 years (range 26–76). Sixteen (30.2%) patients were males and 37 (69.8%) females. The diagnosis was malignant PEComa in 42 (79.2%) cases out of 53 and epithelioid angiomyolipoma in 11 (20.8%).

Median follow-up was 32.5 months [interquartile range (IQR), 9.1–58.8]. Twenty-four patients (45.3%) received only one line of treatment, 14 (26.4%) two lines, 7 (13.2%) three lines, and other 8 (15.1%) received four or more lines. OS of the overall population starting from the first medical therapy was 30.1 months (IQR, 15.4-nr; Fig. 1). Clinical outcomes and specifics for each treatment category are summarized in Table 2 and Supplementary Table S1.

Twenty-three patients treated with anthracycline-based chemotherapy were identified, and all were evaluable for response. Fifteen patients (65.2%) received anthracycline with ifosfamide (AI), 2 (8.7%) received anthracycline combined with dacarbazine, 1 (4.4%) patient received doxorubicin and olaratumab, and 5 (21.7%) received single-agent anthracycline. Fourteen (60.9%) patients were treated with an anthracycline as first-line therapy, 5 (21.7%) in second-line, and 4 (17.4%) in third- or more lines. At the time of analysis, 1 patient continued on treatment. Three (13%) patients had a PR, 10 (43.5%) had stable disease (SD), and 10 (43.5%) had PD. The median PFS was 3.2 months (IQR, 1.8–14; Fig. 2A).

Fifteen patients were treated with gemcitabine-based regimens and all were evaluable for response. Nine (60%) patients were treated with gemcitabine as first-line therapy, 3 (20%) in second-line, 3 (20%) in third-line. At the time of analysis, all patients completed their treatment. Three (20%) had a PR, 2 (13.3%) SD, and 10 (66.7%) PD. The median PFS was 3.4 (IQR, 2.3–4.9) months (Fig. 2B).

Twelve patients received antiangiogenic agents and were available for response. Nine patients (75%) received pazopanib, 2 (16.7%) received sorafenib, 1 (8.3%) patient received sunitinib. Three (25%) patients were treated in second-line, 4 (33.3%) in third-line, and 5 (41.7%) in fourth- or more lines. At the time of analysis, all patients had completed antiangiogenic therapy. Treatment was discontinued for PD in all patients but one (toxicity). One patient (8.3%) had a PR (to sorafenib), 9 (75%) had SD to antiangiogenic therapy, and 2 (16.7) had PD. Prolonged disease stabilization was observed in 2 patients, one with pazopanib and the other with sorafenib, both lasting 12 months.

The median PFS was 5.4 months (IQR, 2.5–9.6; Fig. 2C).

Supplementary Fig. S1 in the Supplement shows a tissue response in an abdominal mass after 1 month of pazopanib as fourth-line therapy in a woman with uterine PEComa. The patient had a complicated response so then underwent surgery. Unfortunately, the patients died 2 months later for a progressive deterioration of general conditions. This patient did not respond to anthracycline or trabectedin but had a PR to gemcitabine lasting 3 months. Patient did not receive mTOR inhibitors as the initial diagnosis was leiomyosarcoma.

Forty patients were treated with mTOR inhibitors and all but one were evaluable for response. Thirty-two patients (80%) received sirolimus, 5 (12.5%) received everolimus, 3 (7.5%) patients received temsirolimus. Thirty (75%) patients were treated in first-line setting, 8 (20%) in second-line, 2 (5%) in third-line. At the time of the last follow-up, 13 (32.5%) patients were still on treatment. Sirolimus was given orally at a dose between 1 and 10 mg daily according to plasma drug level (optimal value between 14–20 ng/mL). Temsirolimus was given at 25 mg intravenous weekly. Everolimus was given orally at 5 mg daily.

One patient (2.6%) had a complete response (CR), 15 (38.5%) PR, 14 (35.9%) SD, and 9 (23%) PD. Eleven (28.2%) had a long-lasting response, of more than 1 year. The median PFS was 9 months (IQR, 4.4–39.4; Fig. 2D). Notably, median PFS of responding patients was 15.4 months (IQR, 8.5–47.3). One female patient with a retroperitoneal PEComa and metastases in the lung was treated with sirolimus, but treatment was stopped after 19 months because of pneumonitis. She progressed after 2 months and responded to sirolimus rechallenge, with a durable response of 12 months prior to experiencing a further disease progression. At the time of such progression, the plasma value of sirolimus was 5.5 ng/mL. The dose of sirolimus was increased from 5 mg to 8 mg. The plasma value increased to 16 ng/dL and the disease was stabilized for over 6 months.

Another patient with hepatic metastases from visceral PEComa, underwent surgery of all hepatic lesions after a PR to sirolimus (Supplementary Fig. S2). The pathology report highlighted a pathologic CR defined as the absence of residual tumor cells, replaced by fibromyxoid stroma, on hematoxylin and eosin evaluation in the entire resected specimen. The patient did not receive mTOR inhibition after surgery and progressed 3 months later. Sirolimus was resumed at progression and a further PR during over 1 year was observed (still ongoing at time of last follow-up).

In eight (20%) patients treated with mTOR inhibitors, the primary tumor site was the uterus, while in 32 (80%) the primary site was extrauterine. In the group of uterine PEComas, 1 of 8 patients had a PR (12.5%), and 3 (37.5%) had SD. In the group of extrauterine PEComas, 16 of 31 evaluable patients by RECIST had a PR (51.6%) and 10 (32.3%) had stable disease. Notably, we observed a nonsignificant trend, toward a worse PFS for patients with uterine versus extrauterine PEComa [median PFS 6.4 vs. 10.4 months, respectively; HR, 1.51; 95% confidence interval (CI), 0.94–2.43; P = 0.09; Supplementary Fig. S3].

Our results show that standard soft-tissue sarcoma chemotherapy regimens are active only in a small proportion of PEComas, as demonstrated by the relatively short median PFS. Gemcitabine-based regimens [objective response rate (ORR): 20%, median PFS: 3.4 months] seemed to have the same activity of anthracycline-based regimens (ORR: 13%, median PFS: 3.2 months). Antiangiogenic agents can result in disease stabilization in some patients (remarkably, a prolonged disease stabilization lasting 12 months was observed in 2 patients), with a number having density changes/tissue response on imaging, as is often observed with this class of drugs. Indeed, the PFS with antiangiogenic agents (5.4 months) slightly contrasts with the ORR (8.3%). mTOR inhibitors were the most active agents in this histologic subtype, with an ORR of 41% and a PFS of 9 months. Notably, a subset of responding patients experienced a response in excess of 1 year (28.2%). Rechallenge with mTOR inhibitors can also result in new responses in these patients. We saw hints that plasma drug monitoring could be an additional information in the management of patients when using these agents. Prior to stopping therapy, it could be advisable to document the plasma level and increase the dose when necessary. However, published data are contrasting regarding this issue. Indeed, in the paper of Wagner and colleagues (6) one patient of the 3 reported, had a PR lasting 10 months despite a sirolimus level of 9.4 ng/mL. Interestingly, extrauterine PEComas show higher response rate to mTOR inhibitors then uterine PEComas, with a nonsignificant trend toward longer PFS.

To our knowledge, this academic, retrospective, international study is the largest series of systemic therapy for locally advanced/metastatic PEComas. Thus, it may provide a benchmark in terms of tumor response, PFS, and OS for future prospective trials. While published data on the use of chemotherapy and antiangiogenetic agents in PEComa (9, 10) are scant, the activity of mTOR inhibitors has been reported (6). Published reports are limited to few cases (6, 11–14) with the largest case series comprising 10 patients (14). A clinical study of mTOR inhibitors is currently ongoing in the United States (ClinicalTrials.gov Identifier: NCT02494570).

Recently, an attempt to expand the knowledge about the molecular landscape of these tumors was made. A subset of PEComas harboring a translocation in TFE3 gene was reported (15). As the presence of the translocation is mutually exclusive with TSC1-2 mutations, this may help to subclassify PEComas and, in the future, may help to tailor a distinct treatment strategy (16), although the differential prevalence of TFE3 translocations and TSC1-2 mutations in uterine and extrauterine is unknown. These observations if confirmed in a future study, could explain, at least in part, the heterogeneity of responses seen to mTOR inhibitors, with the suggestion of a differential response rate between uterine and extrauterine PEComas to mTOR inhibitors. Moreover, uterine PEComa tumors often have a distinct morphology and IHC (17).

In summary, despite the limitations of a retrospective design, our study provides data for the selection of systemic therapy in patients with advanced/metastatic PEComa. Antiangiogenics and chemotherapy with anthracycline or gemcitabine-based regimens are both options. mTOR inhibitors are the most active agents, with a subgroup of patients having durable benefit. The OS of this population calculated from the time of systemic treatment initiation for advanced disease is approximately 30 months, which is consistent with other metastatic soft-tissue sarcomas (18).

R. Sanfilippo reports receiving commercial research grants from Eli Lilly and PharmaMar. R.L. Jones is a consultant/advisory board member for Adaptimmune, Blueprint, Clinigen, Eisai, Epizyme, Daichii, Deciphera, Immunodesign, Lilly, Merck, Pharmamar, and TRACON. J. Blay is a consultant/advisory board member for Bayer, GlaxoSmithKline, Novartis, and Roche. O. Mir has ownership interest in Amplitude Surgical and Transgene, reports receiving speakers bureau honoraria from Eli Lilly, Roche, Pfizer, and Servier, and is a consultant/advisory board member for Amgen, Bayer, Bristol-Myers Squibb, Eli Lilly, Ipsen, Lundbeck, MSD, Novartis, Pfizer, Roche, Servier, and Vifor Pharma. S. Stacchiotti is a consultant/advisory board member for Eli Lilly and Bayer, and reports receiving commercial research support from Bayer, Novartis, Eli Lilly, and Pfizer. A.P. Dei Tos reports receiving speakers bureau honoraria from Lilly and is a consultant/advisory board member for Roche. P.G. Casali reports receiving speakers bureau honoraria from Eisai, Pfizer, and Eli Lilly, and is a consultant/advisory board member for Bayer, Deciphera, Eli Lilly, and Nektar Therapeutics. No potential conflicts of interest were disclosed by the other authors.

Conception and design: R. Sanfilippo, R.L. Jones, R. Bertulli, S. Stacchiotti, A. Gronchi, A.P. Dei Tos, P.G. Casali

Development of methodology: R. Sanfilippo, R.L. Jones, J.-Y. Blay, P. Collini

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R. Sanfilippo, R.L. Jones, J.-Y. Blay, A. Le Cesne, S. Provenzano, G. Antoniou, O. Mir, E. Fumagalli, R. Bertulli, S. Stacchiotti, F. Grosso, A. Dufresne, N. Hindi, M. Sbaraglia, A. Gronchi, P. Collini, A.P. Dei Tos

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Sanfilippo, R.L. Jones, J.-Y. Blay, A. Le Cesne, S. Provenzano, O. Mir, G. Fucà, E. Fumagalli, P. Collini, A.P. Dei Tos

Writing, review, and/or revision of the manuscript: R. Sanfilippo, R.L. Jones, J.-Y. Blay, A. Le Cesne, S. Provenzano, G. Antoniou, O. Mir, G. Fucà, R. Bertulli, S. Stacchiotti, A. Dufresne, N. Hindi, A. Gronchi, P. Collini, A.P. Dei Tos, P.G. Casali

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R. Sanfilippo, R.L. Jones, J.-Y. Blay, G. Fucà, M. Brahmi

Study supervision: R. Sanfilippo, R.L. Jones, A. Gronchi, P.G. Casali

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.