Purpose:

Doxorubicin is standard therapy for advanced soft-tissue sarcoma (STS) with minimal improvement in efficacy and increased toxicity with addition of other cytotoxic agents. Pembrolizumab monotherapy has demonstrated modest activity and tolerability in previous advanced STS studies. This study combined pembrolizumab with doxorubicin to assess safety and efficacy in frontline and relapsed settings of advanced STS.

Patients and Methods:

This single-center, single-arm, phase II trial enrolled patients with unresectable or metastatic STS with no prior anthracycline therapy. Patients received pembrolizumab 200 mg i.v. and doxorubicin (60 mg/m2 cycle 1 with subsequent escalation to 75 mg/m2 as tolerated). The primary endpoint was safety. Secondary endpoints included overall survival (OS), objective response rate (ORR), and progression-free survival (PFS) based on RECIST v1.1 guidelines.

Results:

Thirty patients were enrolled (53.3% female; median age 61.5 years; 87% previously untreated) with 4 (13.3%) patients continuing treatment. The study met its primary safety endpoint by prespecified Bayesian stopping rules. The majority of grade 3+ treatment-emergent adverse events were hematologic (36.7% 3+ neutropenia). ORR was 36.7% [95% confidence interval (CI), 19.9–56.1%], with documented disease control in 80.0% (95% CI, 61.4–92.3%) of patients. Ten (33.3%) patients achieved partial response, 1 (3.3%) patient achieved complete response, and 13 (43.3%) patients had stable disease. Median PFS and OS were 5.7 months (6-month PFS rate: 44%) and 17 months (12-month OS rate: 62%), respectively. Programmed cell death ligand-1 (PD-L1) expression was associated with improved ORR, but not OS or PFS.

Conclusions:

Combination pembrolizumab and doxorubicin has manageable toxicity and preliminary promising activity in treatment of patients with anthracycline-naive advanced STS.

Translational Relevance

Traditional cytotoxic treatment is only marginally effective for metastatic soft-tissue sarcoma (STS). Based on data from as far back as the 1990s, the upper limit of response to single-agent chemotherapy is no better than 20% to 25%. Although inroads have been made with targeted therapies in select subhistologies of STS, for the large majority of patients without a surgical option single-agent anthracycline remains the first-line standard of care. The role of immunotherapy in the management of STS is not fully defined. This study suggests that the combination of anti–programmed cell death receptor 1 (PD-1)-based immunotherapy with standard-of-care chemotherapy is safe and may provide enhanced efficacy in the treatment of metastatic STS. A larger, randomized trial will be required to validate these results.

Sarcomas are a heterogenous group of malignancies that arise from transformed mesenchymal cells (1, 2). Sarcomas are broadly categorized as either bone or soft-tissue sarcomas (STS), with greater than 50 subtypes described to date (2–4). Patients with metastatic or unresectable STS have a poor prognosis, with median overall survival (OS) of 15 to 18 months (3, 5–9). Treatment options for STS are limited and have relied on a small number of cytotoxic agents for more than three decades.

In the setting of metastatic or unresectable disease, standard therapy usually consists of single-agent doxorubicin or gemcitabine/docetaxel chemotherapy with response rates reported between 10% and 25% (10–13). While adding other chemotherapeutic agents to doxorubicin may improve response rates, it comes at the cost of increased toxicity with no improved survival (14–17). Doxorubicin monotherapy has shown comparable, if not slightly enhanced, efficacy and tolerability in patients with STS compared with other treatment options including epirubicin and ifosfamide monotherapies (18, 19). Novel second-line and beyond agents including pazopanib, trabectedin, and eribulin have limited efficacy with objective response rate (ORR) of less than 10%, progression-free survival (PFS) of 4.2 months or less, and OS of less than 14 months (12, 20–22). Lack of viable options highlights the importance of developing alternative classes of therapy to improve patient outcome.

Immunotherapy in combination with targeted agents, radiation, and chemotherapy may be associated with improved outcomes compared with immunotherapy alone (2). The mechanistic rationale for combination immunotherapy/chemotherapy is not entirely clear but there may be a role for enhanced antigen presentation versus immunotherapy alone. Programmed cell death receptor 1 (PD-1)-targeted immune checkpoint blockade represents a new treatment avenue for advanced sarcoma (23–27). PD-1 is a cell-surface protein promoting self-tolerance and immune suppression. Tumor cells expressing programmed death ligand 1 (PD-L1) inhibit antitumor responses by binding to PD-1 receptor on activated cytotoxic T cells (7). Some studies have shown that increased PD-1 expression on tumor-infiltrating lymphocytes (TIL) and increased PD-L1 expression on STS cells portends a worse prognosis (28). While CD8+ TILs are not definitively prognostic in sarcoma, they are associated with improved outcomes to immunotherapy (29). The phase II SARC028 study using PD-1 inhibitor, pembrolizumab, to treat patients with unresectable or metastatic sarcoma demonstrated early promising results with an ORR of 18% (23). Expansion cohorts of this study demonstrated a 23% overall response rate for undifferentiated pleomorphic sarcoma and a 10% overall response rate for dedifferentiated liposarcoma (DDLPS; ref. 30). A recent study by Chen and colleagues showed a 14% response rate in undifferentiated pleomorphic sarcoma (UPS)/DDLPS (31). The role of predictive biomarkers in the treatment of STS remains uncertain, but worthy of further inquiry (32). The aim of this study is to evaluate the safety and efficacy of combining pembrolizumab with standard-of-care doxorubicin for treatment of STS not appropriate for curative-intent surgical resection or metastatic STS.

Study design and participant selection

This single-center, single-arm, phase II trial enrolled 30 patients, ages 18 or older, between April 2017 and December 2019 with a histologically confirmed diagnosis of unresectable or metastatic STS not appropriate for surgical therapy and no prior anthracycline or PD-L1 therapy. Patients with Ewing sarcoma, osteosarcoma, chondrosarcoma, Kaposi's sarcoma, gastrointestinal stromal tumors (GIST), clear cell sarcoma, alveolar soft-cell sarcoma, or any other chemotherapy-resistant soft-tissue or bone sarcoma were excluded. Additionally, patients with an immunodeficiency syndrome, human immunodeficiency virus (HIV)/AIDS, or Hepatitis B/C were excluded. Eligible patients had an Eastern Cooperative Oncology Group (ECOG) performance score of 0 or 1 and adequate hematologic, hepatic, and renal values per physician discretion within 10 days prior to initiating treatment. The study protocol was approved by the FDA and Chesapeake Institutional Review Board. The study adhered to Guidelines for Good Clinical Practice and guiding principles laid out in the Declaration of Helsinki. All participants provided written, informed consent. This trial is registered with ClinicalTrials.gov, number NCT03056001.

Procedures

During the initial study period, participants received doxorubicin by bolus i.v. (60 mg/m2 on cycle 1 with escalation to 75 mg/m2 on cycle 2 as tolerated) and pembrolizumab (200 mg) i.v. every 3 weeks (21 days) for 24 months or until documented disease progression or unacceptable toxicity assessed by Common Terminology Criteria for Adverse Events (CTCAE) version 4. Disease was assessed by CT scan performed within 28 days prior to initiation of treatment (baseline). Scans were performed every 6 weeks during study period until first documented disease progression or start of new anticancer therapy. Off study, an every-9-week scanning schedule was allowed. Treatment was permitted up to one cycle past progression if in the opinion of the investigator the patient was having a treatment benefit. For progression on first set of subsequent scans the patient was removed from the study. Day 1 of each cycle was defined as the day pembrolizumab was administered. If pembrolizumab administration was delayed, doxorubicin was administered every 21 days per standard of care. However, if pembrolizumab was discontinued due to toxicity, doxorubicin therapy was also discontinued. In the event doxorubicin was discontinued due to toxicity, patients continued to receive pembrolizumab for up to 2 years. Patients that reached a lifetime cumulative dose for doxorubicin (450 mg/m2) while on combination therapy were able to continue pembrolizumab monotherapy treatment for up to 2 years. Growth-factor support was permitted for any of the treatment cycles, at physician discretion.

Outcomes

The primary endpoint was safety, based on a Bayesian stopping rule, evaluating the rate of severe or life-threatening treatment-emergent adverse events (TEAE; ref. 31). A severe or life-threatening TEAE was defined by: meeting the definition of seriousness, related to pembrolizumab and/or doxorubicin per the sponsor-investigator, and considered to be clinically significant by the sponsor and/or investigator. Secondary endpoints included OS, ORR as determined by RECIST v1.1, and PFS. OS was defined as time from start of protocol treatment to death from any cause or censored at last known date alive if the patient was alive or lost to follow-up at time of analysis. Objective response (OR) was determined for each patient indicating whether the patient achieved a best overall response of complete response (CR) or partial response (PR), per RECIST v1.1 criteria. Confirmation of response on subsequent imaging was not required as response was not the primary endpoint. PFS was defined as time from start of protocol treatment to first occurrence of progressive disease (PD) or death. The date of PD was date of the radiologic assessment that identified RECIST-defined PD or date the clinician made subjective determination of disease progression. If the patient died without documented disease progression, progression date was date of death. For surviving patients who did not have documented disease progression, PFS was censored at date of last radiologic assessment prior to any subsequent anticancer therapy.

For exploratory endpoints, PD-L1 expression and intensity as determined by H-score (0–300), described by Igarashi and colleagues (2016; ref. 32), was categorized, where low was defined as H-score less than 5 and high as H-score of 5 or greater. TIL scoring was blinded and done by a central lab using the Qualtek TIL assessment, a morphologic assessment of the prescence/absence of TILs within tumor nests graded on a scale of 0 to 3 (0–1: no–low TILs, 2: intermediate TILs, 3: high TILs). A cutoff value of 1 and Fisher exact test were used in the statistical analysis of TIL expression.

Statistical analysis

It has previously been reported that the rate of any severe or life-threatening toxicity of doxorubicin and dacarbazine given every 3 weeks is 55% (31). The Bayesian stopping rule would have held enrollment if posterior probability of severe or life-threatening toxicity rate exceeding 0.55 was 75% or higher. The prior for this monitoring rule is beta (9, 11), therefore the prior assumption for incidence of severe or life-threatening toxicity in patients is 0.55 and there is a 90% probability that this proportion is between 0.368 and 0.726. The stopping-rule boundaries are given in Supplementary Table S1. The study would be determined to be successful if the prespecified enrollment target of 30 patients was completed and the incidence of stopping-rule events (as defined above) in treated patients never exceeded the stopping-rule boundaries indicated in Supplementary Table S1. Response rates were calculated with corresponding 95% confidence intervals (CI) based on the Clopper–Pearson method. OS and PFS distributions were estimated using Kaplan–Meier techniques, with medians and landmarks estimated from Kaplan–Meier curves along with 95% CIs. All efficacy analyses were on the intent-to-treat population (those patients who were enrolled on the study). Exploratory analysis of the association of response with correlative PD-L1 H-score data was evaluated with a Fisher exact test. Additionally, Fisher exact tests and logistic regression models were utilized to evaluate patient demographics and disease characteristics for association with response or disease control. To ensure a fully validated dataset for data disclosure, the cutoff date for this report was April 3, 2020.

Thirty (30) patients with histologically confirmed unresectable or metastatic STS without prior anthracycline therapy were enrolled in this phase II trial with 4 patients (13.3%) remaining on combination doxorubicin and pembrolizumab therapy as of data cut-off date. Most patients (86.7%) received combination doxorubicin and pembrolizumab therapy as first-line systemic treatment. The most common histologies were leiomyosarcoma (LMS; 33.3%), DDLPS (23.3%), and UPS (13.3%) with a median follow-up time of 10.4 months [interquartile range (IQR) 5.3–17.7; Table 1]. The study included 16 female patients (53.3%) with an overall median age at diagnosis of 61.5 years (IQR 52–70), including a third of patients who were 70 years or older. Demographic and clinical data including disease-related factors are listed in Table 1.

Table 1.

Demographics and characteristics of patients with STS.

N = 30%
Gender 
 Male 14 46.7% 
 Female 16 53.3% 
Age at consent 
 18–29 years 6.7% 
 30–49 years 13.3% 
 50–69 years 14 46.7% 
 70+ years 10 33.3% 
Race 
 African American 13.3% 
 Caucasian 23 76.7% 
 Unknown/not reported 10.0% 
Ethnicity 
 Hispanic or Latino 10.0% 
 Non-Hispanic or Latino 26 86.7% 
 Unknown/not reported 3.3% 
Primary tumor site 
 Extremity – arm 6.7% 
 Extremity – leg 26.7% 
 Retroperitoneal/abdomen 13 43.3% 
 Other 23.3% 
Histology 
 Liposarcoma 23.3% 
 Leiomyosarcoma 10 33.3% 
 Synovial sarcoma 3.3% 
 UPS 13.3% 
 Angiosarcoma 6.7% 
 Rhabdomyosarcoma 3.3% 
 Epithelioid angiosarcoma 3.3% 
 Fibromyxoid sarcoma/sclerosing epithelioid fibrosarcoma 3.3% 
 Malignant fibrous histiocytoma 3.3% 
 Spindle-cell solitary fibrous tumor 3.3% 
 Extraskeletal osteosarcoma 3.3% 
Stage (at first histologic confirmation of disease) 
 I 13.3% 
 IIA 3.3% 
 IIB 13.3% 
 III 30.0% 
 IV 11 36.7% 
 Unknown 3.3% 
Histologic grade (at first histologic confirmation of disease) 
 G1 (Low, well differentiated) 13.3% 
 G2 (Intermediate, moderately differentiated) 13.3% 
 G3 (High, poorly differentiated) 20 66.7% 
 Unknown/cannot be assessed 6.7% 
Metastatic site 
 Liver 20.0% 
 Lung 23.3% 
 Lymph node 6.7% 
 Other 14 46.7% 
 No metastatic disease 3.3% 
Prior lines of systemic treatment 
 0 26 86.7% 
 1 3.3% 
 2 10.0% 
N = 30%
Gender 
 Male 14 46.7% 
 Female 16 53.3% 
Age at consent 
 18–29 years 6.7% 
 30–49 years 13.3% 
 50–69 years 14 46.7% 
 70+ years 10 33.3% 
Race 
 African American 13.3% 
 Caucasian 23 76.7% 
 Unknown/not reported 10.0% 
Ethnicity 
 Hispanic or Latino 10.0% 
 Non-Hispanic or Latino 26 86.7% 
 Unknown/not reported 3.3% 
Primary tumor site 
 Extremity – arm 6.7% 
 Extremity – leg 26.7% 
 Retroperitoneal/abdomen 13 43.3% 
 Other 23.3% 
Histology 
 Liposarcoma 23.3% 
 Leiomyosarcoma 10 33.3% 
 Synovial sarcoma 3.3% 
 UPS 13.3% 
 Angiosarcoma 6.7% 
 Rhabdomyosarcoma 3.3% 
 Epithelioid angiosarcoma 3.3% 
 Fibromyxoid sarcoma/sclerosing epithelioid fibrosarcoma 3.3% 
 Malignant fibrous histiocytoma 3.3% 
 Spindle-cell solitary fibrous tumor 3.3% 
 Extraskeletal osteosarcoma 3.3% 
Stage (at first histologic confirmation of disease) 
 I 13.3% 
 IIA 3.3% 
 IIB 13.3% 
 III 30.0% 
 IV 11 36.7% 
 Unknown 3.3% 
Histologic grade (at first histologic confirmation of disease) 
 G1 (Low, well differentiated) 13.3% 
 G2 (Intermediate, moderately differentiated) 13.3% 
 G3 (High, poorly differentiated) 20 66.7% 
 Unknown/cannot be assessed 6.7% 
Metastatic site 
 Liver 20.0% 
 Lung 23.3% 
 Lymph node 6.7% 
 Other 14 46.7% 
 No metastatic disease 3.3% 
Prior lines of systemic treatment 
 0 26 86.7% 
 1 3.3% 
 2 10.0% 

Safety and TEAEs

As of the data cutoff date, median number of pembrolizumab doses delivered was 6.5 (range 1–28) and doxorubicin was 6 (range 1–7). Twenty-one of 30 patients (70%) received at least 1 dose of doxorubicin at 75 mg/m2. The large majority (96.7%) of patients received prophylactic growth factor. Overall, the study did not meet criteria for stopping enrollment early (31); however, 1 patient did experience a stopping-rule event, Grade 3 autoimmune hepatitis, believed to be attributed to both pembrolizumab and doxorubicin (Supplementary Table S1). The most common Grade 3 or higher TEAEs were hematologic and included decreased neutrophil and white blood cell count (36.7% each) and anemia (26.7%; Table 2). Additionally, 11 patients (36.7%) experienced at least one Grade 3 or higher TEAE specifically attributed to pembrolizumab, with the most common including arthralgia (10.0%), fatigue (6.7%), autoimmune disorder (6.7%), and increased lipase (6.7%; Table 3). Seventeen patients (56.7%) had at least one serious adverse event (SAE), with the most frequent being febrile neutropenia (5%), nausea (4%), vomiting (3%), and pulmonary infection (3%). Treatment was discontinued in 5 patients (17%) due to Grade 3 or higher pembrolizumab toxicity. Three of these patients developed joint swelling and redness requiring steroids which were administered on a standard schedule. One of the 3 developed prominent synovitis/myositis (with CPK elevation). One patient had an auto-immune hepatitis and 1 patient possible auto-immune nephritis. All patients with pembrolizumab-related TEAEs responded to steroids, though a protracted course of more than a month was required in some patients. All TEAEs are included in Supplementary Tables S2–S4. Two patients died prior to their first scheduled imaging assessment. They were included in the intention-to-treat analysis. As of data cut-off, half (15/30) of the patients had died, the large majority (13/15) while in follow-up, 14 from their malignancy, and one from unknown causes. No grade 5 treatment-related toxicities were observed.

Table 2.

Most commona grade 3+ TEAEs, per CTCAE Version 4 criteria.

Preferred termN%
Neutrophil count decreased 11 36.7% 
White blood cell decreased 11 36.7% 
Anemia 26.7% 
Febrile neutropenia 16.7% 
Arthralgia 13.3% 
Lymphocyte count decreased 13.3% 
Nausea 13.3% 
Fatigue 10.0% 
Hyponatremia 10.0% 
Vomiting 10.0% 
Lung infection 10.0% 
Generalized muscle weakness 10.0% 
Preferred termN%
Neutrophil count decreased 11 36.7% 
White blood cell decreased 11 36.7% 
Anemia 26.7% 
Febrile neutropenia 16.7% 
Arthralgia 13.3% 
Lymphocyte count decreased 13.3% 
Nausea 13.3% 
Fatigue 10.0% 
Hyponatremia 10.0% 
Vomiting 10.0% 
Lung infection 10.0% 
Generalized muscle weakness 10.0% 

aExperienced by 3 or more patients.

Table 3.

Grade 3+ AEs attributeda to pembrolizumab.

Preferred termN%
At least one grade 3+ AE 11 36.7% 
Arthralgia 10.0% 
Fatigue 6.7% 
Lipase increased 6.7% 
Autoimmune disorder 6.7% 
Alanine aminotransferase increased 3.3% 
Arthritis 3.3% 
Aspartate aminotransferase increased 3.3% 
Blood bilirubin increased 3.3% 
Colitis 3.3% 
Gastritis 3.3% 
Hyperglycemia 3.3% 
Hypokalemia 3.3% 
Serum amylase increased 3.3% 
Generalized muscle weakness 3.3% 
Preferred termN%
At least one grade 3+ AE 11 36.7% 
Arthralgia 10.0% 
Fatigue 6.7% 
Lipase increased 6.7% 
Autoimmune disorder 6.7% 
Alanine aminotransferase increased 3.3% 
Arthritis 3.3% 
Aspartate aminotransferase increased 3.3% 
Blood bilirubin increased 3.3% 
Colitis 3.3% 
Gastritis 3.3% 
Hyperglycemia 3.3% 
Hypokalemia 3.3% 
Serum amylase increased 3.3% 
Generalized muscle weakness 3.3% 

aAttributed: possible, probable, or definite.

Response rates and outcomes

All patients had measurable disease at baseline and were evaluable for ORR. Eleven participants (36.7%, 95% CI, 19.9%–56.1%) achieved an OR, with 1 participant (3.3%, 95% CI, 0.1%–17.2%) achieving a CR. Stable disease (SD) as the best response to treatment was noted in 13 patients (43.3%, 95% CI, 25.5%–62.6%), while PD was seen in 6 patients (20.0%, 95% CI, 7.7%–38.6%), including 2 patients who expired prior to obtaining radiographic scans postbaseline. Best response to treatment according to STS histologic subtype is shown in Table 4. All 4 patients with UPS (100%), 4 of 10 patients with leiomyosarcoma (40%), and 2 of 7 patients with liposarcoma (28.6%) experienced an OR. No significant associations were identified between patient demographics or disease characteristics (age, gender, race, primary tumor site, stage at initial diagnosis, histologic grade at diagnosis, and metastatic disease site) and treatment response.

Table 4.

Response rates and best response in 30 patients with STS by histologic subtype.

HistologyORCRPRSDPD
Liposarcoma (N = 7) 2 (28.6%) 1 (14.3%) 1 (14.3%) 2 (28.6%) 3 (42.9%) 
Leiomyosarcoma (N = 10) 4 (40.0%) 0 (0.0%) 4 (40.0%) 6 (60.0%) 0 (0.0%) 
Synovial sarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
UPS (N = 4) 4 (100.0%) 0 (0.0%) 4 (100.0%) 0 (0.0%) 0 (0.0%) 
Angiosarcoma (N = 2) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (50.0%) 1 (50.0%) 
Rhabdomyosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Epithelioid angiosarcoma (N = 1) 1 (100.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 0 (0.0%) 
Fibromyxoid sarcoma/sclerosing epithelioid fibrosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Malignant fibrous histiocytoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Spindle-cell solitary fibrous tumor (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 
Extraskeletal osteosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 
Overall 11 (36.7%) 1 (3.3%) 10 (33.3%) 13 (43.3%) 6 (20.0%) 
HistologyORCRPRSDPD
Liposarcoma (N = 7) 2 (28.6%) 1 (14.3%) 1 (14.3%) 2 (28.6%) 3 (42.9%) 
Leiomyosarcoma (N = 10) 4 (40.0%) 0 (0.0%) 4 (40.0%) 6 (60.0%) 0 (0.0%) 
Synovial sarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
UPS (N = 4) 4 (100.0%) 0 (0.0%) 4 (100.0%) 0 (0.0%) 0 (0.0%) 
Angiosarcoma (N = 2) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (50.0%) 1 (50.0%) 
Rhabdomyosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Epithelioid angiosarcoma (N = 1) 1 (100.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 0 (0.0%) 
Fibromyxoid sarcoma/sclerosing epithelioid fibrosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Malignant fibrous histiocytoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 0 (0.0%) 
Spindle-cell solitary fibrous tumor (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 
Extraskeletal osteosarcoma (N = 1) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100.0%) 
Overall 11 (36.7%) 1 (3.3%) 10 (33.3%) 13 (43.3%) 6 (20.0%) 

The best change in target-lesion size from baseline, categorized by best response, is plotted in Fig. 1A. For this analysis, 28 of 30 (93.3%) patients were evaluable, as 2 patients expired without obtaining radiographic scans after baseline. Decreased tumor burden from baseline was observed in 18 (60.0%) patients. Among patients who achieved an OR, the average reduction in tumor burden was 62.6%. Patients with a best response of SD had an average reduction in tumor burden of 8.0%, while those with PD had an average 16.9% increase in tumor burden. Note 2 patients with best response of PD had progressed due to appearance of new lesions rather than an increase in target-lesion size from baseline scans (one of whom had >30% reduction in size of target lesions). The clinical courses of enrolled patients, including duration of response and disease control, are depicted in Fig. 1B. At time of analysis, 8 (26.7%) patients showed continued disease control or response to combination therapy with a median duration of disease control of 7.1 months (95% CI, 5.5–9.9). Of note, 1 patient with liposarcoma showed a PR to treatment at 7 months with CR at 18 months.

Figure 1.

A, Waterfall plot of best change in tumor burden from baseline. In 2 patients, denoted by a blue star, disease progression was due to presence of new lesions and not by an increase in size of the target lesion from baseline. Postbaseline scans were not available for 2 patients due to death prior to second set of scans (N = 28). B, Duration of response and disease control status (N = 30).

Figure 1.

A, Waterfall plot of best change in tumor burden from baseline. In 2 patients, denoted by a blue star, disease progression was due to presence of new lesions and not by an increase in size of the target lesion from baseline. Postbaseline scans were not available for 2 patients due to death prior to second set of scans (N = 28). B, Duration of response and disease control status (N = 30).

Close modal

OS and PFS for combination doxorubicin and pembrolizumab are depicted in Fig. 2A and B, respectively. The median survival was 17.0 (95% CI, 9.9–not reached) months, with 12-month OS rate of 62.4% (95% CI, 41.1–77.9%; Fig. 2A). The median PFS was 5.7 (95% CI, 4.1–8.9) months, with 6-month PFS rate of 44.4% (95% CI, 25.2–62.0%), and 12-month PFS of 20.3% (95% CI, 6.8%–38.8%; Fig. 2B).

Figure 2.

Kaplan–Meier curves for OS (A) and PFS (B; N = 30).

Figure 2.

Kaplan–Meier curves for OS (A) and PFS (B; N = 30).

Close modal

Correlative studies

We performed an exploratory analysis of PD-L1 expression (H-score) and TILs. IHC for exploratory endpoints was available for 29 of 30 patients (27 from archived tissue and 2 from on-study biopsies). Thirty-eight percent of evaluable patients had PD-L1 H-score ≥ 5% (55% PDL H-score ≥1%) and had a greater ORR (63.6%) than those with an H-score of <5% (22.2%; P = 0.048). While PD-L1 expression was not found to be associated with PFS or OS, OS analysis is still immature. TILs were not found to be associated with ORR, PFS, or OS alone or in combination with PD-L1 expression.

This study of a rare tumor met its primary safety endpoint. The combination of pembrolizumab and doxorubicin demonstrated similar adverse events (AE) to those observed in other clinical trials without additional concerning safety signals. Grade 3/4 neutropenia occurred in 36.7% of patients with grade 3/4 febrile neutropenia in 16.7%, similar to that seen with single-agent doxorubicin in the 2014 trial by Judson and colleagues (37% and 13%, respectively; ref. 14). Grade 3/4 arthralgia was noted in 13% of our patients which is not commonly reported with doxorubicin monotherapy but has been noted in other immunotherapy trials. In SARC028 trial, 10% of treated patients with STS had grade 3/4 connective tissue/musculoskeletal AEs (23).

The combination of pembrolizumab with doxorubicin demonstrated an ORR of 36.7% compared with previously reported pembrolizumab alone (18%) and single-agent doxorubicin (13%; ref. 23). Comparison with historical series of single-agent doxorubicin is limited given varying dosing schedules of drug and heterogeneity of sarcoma subhistologies studied. A recently published phase I/II trial by Pollack and colleagues evaluated doxorubicin and pembrolizumab in 37 patients with STS (33). Approximately one quarter of enrolled patients received combination therapy as second-line or later therapy. While this study had a lower ORR (19%) than our study, the observed OS was longer at 27 months. Median PFS was 8.1 months in the Pollack study compared with 5.7 months in our study. No musculoskeletal Grade 3 toxicity was noted in the Pollack trial and there were lower rates of Grade 3 neutropenia and neutropenic fever. Differences in outcomes may be due to variability in sarcoma subtypes enrolled, disproportionate proportion of patients receiving it as first lines of therapy, and doxorubicin doses received. In the SARC028 trial, patients with LMS were more inert to the effects of immunotherapy whereas DDLPS and UPS appeared to be more sensitive (23). While the most common subtype enrolled in our trial and the phase I/II trial by Pollack and colleagues was LMS (30% and 33%, respectively), our trial had more patients with DDLPS and UPS (23% and 13.3%, respectively) compared with 11% and 8%, respectively (14, 33). The overall response rate to the combination treatment in this trial compares favorably with traditional aggressive–combination chemotherapy for STSs with doxorubicin and ifosfamide (14).

Sarcomas display a spectrum of immunogenicity with varied levels of tumor-associated inflammation and immunogenicity (34). Despite this fact, most patients benefitted from combination treatment with a disease control rate of 80%, although response rates differed across the various STS histologies. Combination therapy resulted in an ORR of 37% in our trial compared with the SARC028 study which demonstrated an 18% overall response rate with single-agent pembrolizumab to treat relapsed soft-tissue and bone sarcomas (23). Similar to the response seen in patients with UPS receiving pembrolizumab monotherapy in the SARC028 trial, all 4 patients with UPS in this trial responded to doxorubicin and pembrolizumab (23).

Understanding the distinct immune and molecular biology of STS subtypes is crucial to identify who will optimally respond to treatment. Prior studies have shown mixed results in predicting the benefit from PD-1 blockade (24). STSs in general have been described as minimally immunogenic with lower PD-L1 expression and lower tumor mutational burden (25, 34). Of note, some patients in this trial demonstrate a durable response. One patient with LPS achieved a CR and has an ongoing response of 22 months; another with UPS has a sustained partial response after 21 months of treatment. A recent phase II trial by Wilky and colleagues combined axitinib and pembrolizumab for advanced STS (25). The combination showed modest PFS and response with some long-term responses in patients with alveolar soft-part sarcoma. Durable immune response may require the combination of immune-checkpoint inhibitors with chemotherapy, radiation, adoptive cellular therapies, or metabolic therapies (25). Combinations of immunotherapies, for example as used in a phase II study of Talimogene Laherparepec plus pembrolizumab by Kelly and colleagues may be needed for depth and duration of response (26).

The tumor-immune microenvironment and its interplay with immunotherapy response continues to be an ongoing area of research. The heterogeneity among tumor subtypes enrolled in sarcoma trials can pose a challenge in determining correlations. While previous STS studies have shown variable results in correlation between PD-L1 status and immunotherapy response, ours demonstrated a significant correlation. Patients who had tumors with PD-L1 H-score ≥ 5% had a 3-times greater response rate than those with PD-L1 H-score < 5%. There was no correlation seen with response rate and presence of TILs (given the limitation that TIL phenotyping was not rigorously assessed).

Due to heterogeneity of tumor subtypes and limited numbers of patients to enroll, study of these rare tumors is challenging but desperately needed. The results of this study are encouraging and suggest a potential benefit for the combination of anthracycline and checkpoint inhibitors in treatment of STS. The conclusions are preliminary and limited due to small sample size, lack of randomization, single-institution setting, and lack of central radiology review. Nevertheless, taken in conjunction with other early studies of immunotherapy (SARC028; refs. 23, 30), they provide a rationale for a randomized phase III trial using a standard treatment arm of single-agent doxorubicin versus combination doxorubicin/PD-1 inhibitor. Based on what appears to be a favorable sensitivity of UPS and DDLPS (and perhaps LMS) to PD1-based treatment chemoimmunotherapy, an initial study restricted to those subhistologies should be considered. A larger, randomized study is needed to confirm the clinical efficacy of combination pembrolizumab and doxorubicin observed in this study, and to clarify the potential impact of this combination treatment on survival of patients with STS.

Conclusions

This phase II study found the combination of pembrolizumab and doxorubicin to be safe and feasible. In addition, the results suggest enhanced clinical efficacy of the combination of pembrolizumab and doxorubicin versus doxorubicin alone. Like other STS trials, heterogeneity in subtypes enrolled and relatively small numbers of patients need to be considered when evaluating results. There were some exceptional, long-term responders in this study, suggesting a possibly important but not yet fully defined role for immunotherapy in management of advanced STS.

J.T. Symanowski reports personal fees from Immatics and CARsgen, as well as personal fees from Eli Lilly and Company outside the submitted work. E.S. Kim reports personal fees from AstraZeneca, Boehringer Ingelheim, Genentech, Mirati, and Eli Lilly and Company outside the submitted work. No disclosures were reported by the other authors.

M.B. Livingston: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M.H. Jagosky: Conceptualization, formal analysis, methodology, writing–original draft, writing–review and editing. M.M. Robinson: Data curation, formal analysis, validation. W.A. Ahrens: Conceptualization, resources. J.H. Benbow: Conceptualization, formal analysis, writing–original draft. C.J. Farhangfar: Conceptualization, formal analysis, writing–review and editing. D.M. Foureau: Formal analysis, writing–review and editing. D.M. Maxwell: Resources, data curation. E.A. Baldrige: Conceptualization, formal analysis, writing–original draft. X. Begic: Data curation, project administration. J.T. Symanowski: Conceptualization, formal analysis. N.M. Steuerwald: Conceptualization, formal analysis. C.J. Anderson: Conceptualization, formal analysis. J.C. Patt: Conceptualization, validation. J.S. Kneisl: Conceptualization, validation. E.S. Kim: Conceptualization, resources, formal analysis, funding acquisition, methodology, writing–original draft, writing–review and editing.

We gratefully acknowledge The Paula Takacs Foundation for Sarcoma Research and Sue Udelson for their support of the Levine Cancer Institute sarcoma research program. The study drug was provided by Merck (Merck & Co.). Partial funding was provided by Merck and Paula Takacs Foundation.

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.

1.
Canter
RJ
,
Beal
S
,
Borys
D
,
Martinez
SR
,
Bold
RJ
,
Robbins
AS
. 
Interaction of histologic subtype and histologic grade in predicting survival for soft-tissue sarcomas
.
J Am Coll Surg
2010
;
210
:
191
8
.
2.
Hall
F
,
Villalobos
V
,
Wilky
B
. 
Future directions in soft tissue sarcoma treatment
.
Curr Probl Cancer
2019
;
43
:
300
7
.
3.
Liegl-Atzwanger
B
. 
The role of pathology in sarcoma
.
Memo Maga Eur Med Oncol
2020
;
13
:
159
63
.
4.
Katz
D
,
Palmerini
E
,
Pollack
S
. 
More than 50 subtypes of soft tissue sarcoma: paving the path for histology-driven treatments
.
American Society of Clinical Oncology Educational Book
2018
;
38
:
925
38
. Available from: https://ascopubs.org/doi/pdf/10.1200/PO.18.00070.
5.
American Cancer Society
. 
Key statistics for soft tissue sarcomas
. 
2021
. Available from: https://www.cancer.org/cancer/soft-tissue-sarcoma/about/key-statistics.html#references.
6.
Dong
H
,
Strome
SE
,
Salomao
DR
,
Tamura
H
,
Hirano
F
,
Flies
DB
, et al
Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion
.
Nat Med
2002
;
8
:
793
800
.
7.
Disis
ML
. 
Immune regulation of cancer
.
J Clin Oncol
2010
;
28
:
4531
8
.
8.
Ducimetiere
F
,
Lurkin
A
,
Ranchere-Vince
D
,
Decouvelaere
AV
,
Peoc'h
M
,
Istier
L
, et al
Incidence of sarcoma histotypes and molecular subtypes in a prospective epidemiological study with central pathology review and molecular testing
.
PLoS One
2011
;
6
:
e20294
.
9.
Blay
JY
,
Sleijfer
S
,
Schoffski
P
,
Kawai
A
,
Brodowicz
T
,
Demetri
GD
, et al
International expert opinion on patient-tailored management of soft tissue sarcomas
.
Eur J Cancer
2014
;
50
:
679
89
.
10.
Benjamin
RS
,
Wiernik
PH
,
Bachur
NR
. 
Adriamycin: a new effective agent in the therapy of disseminated sarcomas
.
Med Pediatr Oncol
1975
;
1
:
63
76
.
11.
Thorn
CF
,
Oshiro
C
,
Marsh
S
,
Hernandez-Boussard
T
,
McLeod
H
,
Klein
TE
, et al
Doxorubicin pathways: pharmacodynamics and adverse effects
.
Pharmacogenet Genomics
2011
;
21
:
440
6
.
12.
Linch
M
,
Miah
AB
,
Thway
K
,
Judson
IR
,
Benson
C
. 
Systemic treatment of soft-tissue sarcoma-gold standard and novel therapies
.
Nat Rev Clin Oncol
2014
;
11
:
187
202
.
13.
Seddon
B
,
Strauss
SJ
,
Whelan
J
,
Leahy
M
,
Woll
PJ
,
Cowie
F
, et al
Gemcitabine and docetaxel versus doxorubicin as first-line treatment in previously untreated advanced unresectable or metastatic soft-tissue sarcomas (GeDDiS): a randomised controlled phase 3 trial
.
Lancet Oncol
2017
;
18
:
1397
410
.
14.
Judson
I
,
Verweij
J
,
Gelderblom
H
,
Hartmann
JT
,
Schoffski
P
,
Blay
JY
, et al
Doxorubicin alone versus intensified doxorubicin plus ifosfamide for first-line treatment of advanced or metastatic soft-tissue sarcoma: a randomised controlled phase 3 trial
.
Lancet Oncol
2014
;
15
:
415
23
.
15.
D'Ambrosio
L
,
Touati
N
,
Blay
JY
,
Grignani
G
,
Flippot
R
,
Czarnecka
AM
, et al
Doxorubicin plus dacarbazine, doxorubicin plus ifosfamide, or doxorubicin alone as a first-line treatment for advanced leiomyosarcoma: A propensity score matching analysis from the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group
.
Cancer
2020
;
126
:
2637
47
.
16.
Ryan
CW
,
Merimsky
O
,
Agulnik
M
,
Blay
JY
,
Schuetze
SM
,
Van Tine
BA
, et al
PICASSO III: A phase III, placebo-controlled study of doxorubicin with or without palifosfamide in patients with metastatic soft tissue sarcoma
.
J Clin Oncol
2016
;
34
:
3898
905
.
17.
Tap
WD
,
Papai
Z
,
Van Tine
BA
,
Attia
S
,
Ganjoo
KN
,
Jones
RL
, et al
Doxorubicin plus evofosfamide versus doxorubicin alone in locally advanced, unresectable or metastatic soft-tissue sarcoma (TH CR-406/SARC021): an international, multicentre, open-label, randomised phase 3 trial
.
Lancet Oncol
2017
;
18
:
1089
103
.
18.
Lorigan
P
,
Verweij
J
,
Papai
Z
,
Rodenhuis
S
,
Le Cesne
A
,
Leahy
MG
, et al
Phase III trial of two investigational schedules of ifosfamide compared with standard-dose doxorubicin in advanced or metastatic soft tissue sarcoma: a European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group Study
.
J Clin Oncol
2007
;
25
:
3144
50
.
19.
Schoenfeld
DA
,
Rosenbaum
C
,
Horton
J
,
Wolter
JM
,
Falkson
G
,
DeConti
RC
. 
A comparison of adriamycin versus vincristine and adriamycin, and cyclophosphamide versus vincristine, actinomycin-D, and cyclophosphamide for advanced sarcoma
.
Cancer
1982
;
50
:
2757
62
.
20.
van der Graaf
WT
,
Blay
JY
,
Chawla
SP
,
Kim
DW
,
Bui-Nguyen
B
,
Casali
PG
, et al
Pazopanib for metastatic soft-tissue sarcoma (PALETTE): a randomised, double-blind, placebo-controlled phase 3 trial
.
Lancet
2012
;
379
:
1879
86
.
21.
Demetri
GD
,
von Mehren
M
,
Jones
RL
,
Hensley
ML
,
Schuetze
SM
,
Staddon
A
, et al
Efficacy and safety of trabectedin or dacarbazine for metastatic liposarcoma or leiomyosarcoma after failure of conventional chemotherapy: results of a phase III randomized multicenter clinical trial
.
J Clin Oncol
2016
;
34
:
786
93
.
22.
Schoffski
P
,
Ray-Coquard
IL
,
Cioffi
A
,
Bui
NB
,
Bauer
S
,
Hartmann
JT
, et al
Activity of eribulin mesylate in patients with soft-tissue sarcoma: a phase 2 study in four independent histological subtypes
.
Lancet Oncol
2011
;
12
:
1045
52
.
23.
Tawbi
HA
,
Burgess
M
,
Bolejack
V
,
Van Tine
BA
,
Schuetze
SM
,
Hu
J
, et al
Pembrolizumab in advanced soft-tissue sarcoma and bone sarcoma (SARC028): a multicentre, two-cohort, single-arm, open-label, phase 2 trial
.
Lancet Oncol
2017
;
18
:
1493
50
.
24.
Toulmonde
M
,
Penel
N
,
Adam
J
,
Chevreau
C
,
Blay
JY
,
Le Cesne
A
, et al
Use of PD-1 targeting, macrophage infiltration, and IDO pathway activation in sarcomas: a phase 2 clinical trial
.
JAMA Oncol
2018
;
4
:
93
7
.
25.
Wilky
BA
,
Trucco
MM
,
Subhawong
TK
,
Florou
V
,
Park
W
,
Kwon
D
, et al
Axitinib plus pembrolizumab in patients with advanced sarcomas including alveolar soft-part sarcoma: a single-centre, single-arm, phase 2 trial
.
Lancet Oncol
2019
;
20
:
837
48
.
26.
Kelly
CM
,
Antonescu
CR
,
Bowler
T
,
Munhoz
R
,
Chi
P
,
Dickson
MA
, et al
Objective response rate among patients with locally advanced or metastatic sarcoma treated with talimogene laherparepvec in combination with pembrolizumab: a phase 2 clinical trial
.
JAMA Oncol
2020
;
6
:
402
8
.
27.
Maki
RG
,
Jungbluth
AA
,
Gnjatic
S
,
Schwartz
GK
,
D'Adamo
DR
,
Keohan
ML
, et al
A pilot study of anti-CTLA4 antibody ipilimumab in patients with synovial sarcoma
.
Sarcoma
2013
;
2013
:
168145
.
28.
Kim
JR
,
Moon
YJ
,
Kwon
KS
,
Bae
JS
,
Wagle
S
,
Kim
KM
, et al
Tumor infiltrating PD1-positive lymphocytes and the expression of PD-L1 predict poor prognosis of soft tissue sarcomas
.
PLoS One
2013
;
8
:
e82870
.
29.
Sorbye
SW
,
Kilvaer
TK
,
Valkov
A
,
Donnem
T
,
Smeland
E
,
Al-Shibli
K
, et al
Prognostic impact of peritumoral lymphocyte infiltration in soft tissue sarcomas
.
BMC Clin Pathol
2012
;
12
:
5
.
30.
Burgess
MA
,
Bolejack
V
,
Schuetze
S
,
Tine
BAV
,
Attia
S
,
Riedel
RF
, et al
Clinical activity of pembrolizumab (P) in undifferentiated pleomorphic sarcoma (UPS) and dedifferentiated/pleomorphic liposarcoma (LPS): Final results of SARC028 expansion cohorts
.
J Clin Oncol
37
:
15s
, 
2019
(
suppl; abstr 11015
).
31.
Antman
K
,
Crowley
J
,
Balcerzak
SP
,
Rivkin
SE
,
Weiss
GR
,
Elias
A
, et al
An intergroup phase III randomized study of doxorubicin and dacarbazine with or without ifosfamide and mesna in advanced soft tissue and bone sarcomas
.
J Clin Oncol
1993
;
11
:
1276
85
.
32.
Igarashi
T
,
Teramoto
K
,
Ishida
M
,
Hanaoka
J
,
Daigo
Y
. 
Scoring of PD-L1 expression intensity on pulmonary adenocarcinomas and the correlations with clinicopathological factors
.
ESMO Open
2016
;
1
:
e000083
.
33.
Pollack
SM
,
Redman
MW
,
Baker
KK
,
Wagner
MJ
,
Schroeder
BA
,
Loggers
ET
, et al
Assessment of doxorubicin and pembrolizumab in patients with advanced anthracycline-naive sarcoma: A phase 1/2 nonrandomized clinical trial
.
JAMA Oncol
2020
;
6
:
1
5
.
34.
Pollack
SM
,
He
Q
,
Yearley
JH
,
Emerson
R
,
Vignali
M
,
Zhang
Y
, et al
T-cell infiltration and clonality correlate with programmed cell death protein 1 and programmed death-ligand 1 expression in patients with soft tissue sarcomas
.
Cancer
2017
;
123
:
3291
304
.