Abstract

Background

To describe chemotherapy treatment patterns and clinical outcomes in metastatic soft tissue sarcoma (mSTS) patients with favorable response to chemotherapy.

Patients and methods

Multicenter (25) multi-country (9) retrospective chart review of mSTS patients with favorable response to chemotherapy, defined as stable disease or better following four cycles.

Results

Two hundred and thirteen patients (58% female; mean age 54.7 years) received a mean of 2.7 lines of chemotherapy and 5.2 cycles per line. The most common first-line regimens were doxorubicin (34%) and anthracycline plus ifosfamide (30%). Favorable response was achieved by 83% to first-line and 42% and 38% in second- and third-line chemotherapy. The most common reason for chemotherapy discontinuation in lines with a favorable response was reaching a predefined number of cycles in first line (64% of 213) and disease progression in second or later lines (41% of 138). The mean time off chemotherapy was 38.0 weeks after first line, falling to 2.7–6.4 weeks in second or later lines. Median overall and progression-free survival were 23.5 (95% confidence interval 20.5–28.1) and 8.3 (7.4–9.9) months from first favorable response to chemotherapy.

Conclusions

mSTS patients achieving favorable response to chemotherapy have poor outcomes. Additional treatment options are needed.

introduction

Soft-tissue sarcomas (STSs) are malignant tumors of connective tissue, accounting for ∼1% of all human cancers [1]. There are >50 different histological subtypes of STS, with leiomyosarcoma and liposarcoma as the most common subtypes [1]. The annual incidence of STS ranges from 2.4 to 3.6 new cases per 100 000 in population-based studies [2–4]. Approximately 50% of patients diagnosed with STS are diagnosed with or expected to progress to metastatic disease (mSTS) [4, 5].

Surgery is the standard primary management for patients with resectable STS, except for those who present with metastatic disease who may simply receive chemotherapy. Radiotherapy is used routinely as adjuvant treatment for patients with intermediate to high-grade tumors, having been shown to improve local disease control. The value of adjuvant or neoadjuvant chemotherapy remains a subject of considerable debate [6]. Guidelines recommend first-line anthracycline-based (with or without ifosfamide) chemotherapy as the standard of care for mSTS patients, with trabectedin, gemcitabine with or without docetaxel, and dacarbazine reserved for second or later lines of treatment [6–8]. Furthermore, it is now recognized that the choice of chemotherapy may be adapted to the site of the disease and histological or molecular subtype [9]. For example, paclitaxel appears to have specific efficacy in angiosarcomas and gemcitabine, with or without docetaxel, and is active against leiomyosarcoma, especially uterine [10–12]. Trabectedin has a high level of activity in leiomyosarcoma and liposarcoma, in particular myxoid liposarcoma [13, 14].

While there is considerable literature from clinical trials on the outcomes associated with specific treatments in mSTS, there is limited observational data to describe the outcomes of real-life clinical practice in patients with mSTS treated with chemotherapy [15, 16]. The rationale for this study was to understand the treatment patterns and outcomes of mSTS patients with favorable response to chemotherapy across a range of countries.

patients and methods

study design

The SArcoma treatment and Burden of Illness in North America and Europe (SABINE) study was a multi-country, multisite retrospective chart review and cross-sectional patient survey conducted in nine countries: Canada, France, Germany, Italy, The Netherlands, Spain, Sweden, UK, and United States. The chart review portion of SABINE is presented here, while the patient survey is published elsewhere [17]. Patient data were collected from initiation of first-line chemotherapy, between January 2004 and December 2009, to death or end of follow-up.

patients

Patients (age ≥13 years) with a confirmed diagnosis of metastatic STS, with one of the following subtypes: leiomyosarcoma, liposarcoma, synovial sarcoma, or undifferentiated pleomorphic sarcoma/malignant fibrous histiocytoma were included. Patients were also required to have favorable response within at least one line of chemotherapy, with favorable response defined as no disease progression documented by imaging before completion of fourth cycle of chemotherapy and favorable response (complete response, partial response, or stable disease, according to World Health Organisation or RECIST criteria) between start of the third cycle and discontinuation of the line of chemotherapy. Patients participating in the SUCCEED (Sarcoma MultiCenter Clinical Evaluation of the Efficacy of Ridaforolimus) trial were excluded [18].

procedures

The study was approved by the local ethics committee at each site. Data collection forms were designed and collected detailed information on chemotherapy, reasons for discontinuation and modification, use of sarcoma-related imaging, and response status. Data on chemotherapy were collected by line of therapy, including number of cycles, drug name, and dose. For i.v. chemotherapies, the dates and places of administration were also collected. For oral chemotherapies, the start and stop dates and number of doses received were collected. The reasons for regimen discontinuation and the occurrence of and reasons for regimen modification, defined as dosage reduction or discontinuation of drug in a multidrug regimen, were collected within each line of chemotherapy. Sarcoma-related imaging was recorded, including type of imaging test, date imaging carried out, and response (complete response, partial response, stable disease, progressive disease, and not evaluated) for each imaging test.

To better describe the progression and treatment of disease over time, patient follow-up time was divided into mutually exclusive health states based on chemotherapy use and disease progression, within lines of therapy. Three mutually exclusive health states (on chemotherapy, off chemotherapy pre-progression, off chemotherapy post-progression) were defined within four lines of therapy (1, 2, 3, and 4+). The on chemotherapy period started the day of the first dose and ended 3 weeks after the last dose of chemotherapy, with the off chemotherapy period starting the following day and ending the day before the first dose of the next line of therapy. The off-chemotherapy period was divided into two periods—the time periods before and following disease progression (pre- and post-progression), also within lines of therapy. The post-progression period started the date progressive disease was documented from an imaging result. Disease progression status was reset to ‘pre-progression’ with the first dose of a new line of therapy.

statistical analysis

The analysis of chemotherapy treatment patterns was descriptive, with analysis by line of therapy. The percentage of patients exceeding the recommended dose of doxorubicin (450 mg/m2) or epirubicin (900 mg/m2), after which the risk of cardiotoxicity increases, was estimated [8, 19].

The Kaplan–Meier product-limit estimator was used to determine the mean time spent in each state, until death or move to the next health state. In cases where the full survival curve was not observed (the last observation was censored), the curve was extrapolated using an exponential survival function. Mean time spent in each health state (for a patient entering this state) was multiplied by the lifetime probability of the patient entering the health state, yielding the expected time spent in the state. The lifetime probability was, in turn, calculated based on observed transitions between health states. Time-to-event analyses were conducted for overall survival (OS) and progression-free survival (PFS) using the Kaplan–Meier method, from the date first favorable response was attained, and stratified by line of therapy where favorable response was first observed. The date of favorable response was the date when four cycles of chemotherapy were completed. Additional time to event analyses was conducted from initiation of first- and second-line chemotherapy and from diagnosis of metastatic disease to allow for comparison to published literature. Patients who started a subsequent line of chemotherapy before progressing were not censored for PFS.

results

Patient demographic, clinical, and chemotherapy treatment characteristics are shown in Table 1. Of the 213 assessable patients, 42% were male. Patients were treated at centers in France (29%), UK (14%), The Netherlands (14%), United States (13%), Spain (12%), Italy (8%), Germany (7%), Canada (3%), and Sweden (2%). The most frequent subtype was leiomyosarcoma (46%). The most common site of metastasis was lung (48%). The majority of patients (83%) achieved first favorable response during first-line chemotherapy. The percent of patients receiving second, third, and fourth lines of chemotherapy was 76%, 50%, and 25%, respectively. More than half of patients (56%) participated in clinical trials, and 63% were deceased at the time of chart abstraction.

Table 1.

Patient demographic, disease, and treatment characteristics

Characteristics N = 213 
Age at diagnosis of metastatic disease, mean (SD) 54.7 (12.9) 
Female 124 (58.2%) 
Histology  
 Leiomyosarcoma 99 (46.5%) 
 Liposarcoma 55 (25.8%) 
 MFH/PS 34 (16%) 
 Synovial sarcoma 25 (11.7%) 
Grade  
 High 118 (55.4%) 
 Intermediate 56 (26.3%) 
 Low 20 (9.4%) 
 Missing 19 (8.9%) 
Metastatic disease at primary diagnosis 28 (13.1%) 
Patients receiving chemotherapy  
 First line 213 (100%) 
 Second line 161 (75.6%) 
 Third line 107 (50.2%) 
 Fourth line 54 (25.4%) 
 Fifth line 33 (15.5%) 
Line of chemotherapy with first favorable response  
 First line 176 (82.6%) 
 Second line 25 (11.7%) 
 Third line or above 12 (5.6%) 
First favorable response by type  
 Complete response 5 (2.3%) 
 Partial response 37 (17.4%) 
 Stable disease 171 (80.3%) 
Characteristics N = 213 
Age at diagnosis of metastatic disease, mean (SD) 54.7 (12.9) 
Female 124 (58.2%) 
Histology  
 Leiomyosarcoma 99 (46.5%) 
 Liposarcoma 55 (25.8%) 
 MFH/PS 34 (16%) 
 Synovial sarcoma 25 (11.7%) 
Grade  
 High 118 (55.4%) 
 Intermediate 56 (26.3%) 
 Low 20 (9.4%) 
 Missing 19 (8.9%) 
Metastatic disease at primary diagnosis 28 (13.1%) 
Patients receiving chemotherapy  
 First line 213 (100%) 
 Second line 161 (75.6%) 
 Third line 107 (50.2%) 
 Fourth line 54 (25.4%) 
 Fifth line 33 (15.5%) 
Line of chemotherapy with first favorable response  
 First line 176 (82.6%) 
 Second line 25 (11.7%) 
 Third line or above 12 (5.6%) 
First favorable response by type  
 Complete response 5 (2.3%) 
 Partial response 37 (17.4%) 
 Stable disease 171 (80.3%) 

MFH, malignant fibrous histiocytoma; PS, undifferentiated pleomorphic sarcoma.

chemotherapy treatment patterns

Chemotherapy use is presented in Table 2. There were 59 different regimens taken by 213 patients in 585 lines of therapy. On average, patients received 2.7 lines of chemotherapy and 5.2 cycles (range: 1–42) per line. Across all lines of therapy, the most common regimen was doxorubicin (n = 95), followed by trabectedin (n = 85), and docetaxel plus gemcitabine (n = 67).

Table 2.

Chemotherapy regimens taken by at least 15 patients, number of patients, and cycles per patient

  First line Second line Third line Fourth line Fifth line or later All lines 
All regimens N 213 161 107 53 51 585 
Mean (range) 6 (1–42) 4.9 (1–25) 5.2 (1–24) 4.2 (1–20) 4 (1–25) 5.2 (1–42) 
Doxorubicin N (%) 72 (33.8%) 11 (6.8%) 6 (5.6%) 2 (3.8%) 4 (7.8%) 95 (16.2%) 
Mean (range) 5.6 (2–14) 4.5 (1–6) 4.0 (1–7) 2.0 (2–2) 4.0 (1–10) 5.3 (1–14) 
Trabectedin N (%) 10 (4.7%) 23 (14.3%) 26 (24.3%) 15 (28.3%) 11 (21.6%) 85 (14.5%) 
Mean (range) 14.2 (4–42) 7.1 (1–24) 6.3 (1–23) 5.9 (2–20) 4.5 (1–17) 7.1 (1–42) 
GC + Doc N (%) 23 (10.8%) 29 (18.0%) 10 (9.3%) 5 (9.4%) 67 (11.4%) 
Mean (range) 6.7 (3–14) 5.1 (1–12) 9 (3–24) 4.8 (3–10)  6.2 (1–24) 
Doxorubicin + IFO N (%) 45 (21.1%) 8 (5.0%) 1 (0.9%) 54 (9.2%) 
Mean (range) 5.2 (1–7) 3.6 (2–6) 5.0   5.0 (1–7) 
IFO N (%) 2 (0.9%) 27 (16.8%) 9 (8.4%) 4 (7.5%) 3 (5.9%) 45 (7.7%) 
Mean (range) 3.5 (3–4) 3.9 (1–9) 3.1 (1–6) 3.3 (2–6) 4 (2–6) 3.7 (1–9) 
Investigational N (%) 3 (1.4%) 14 (8.7%) 14 (13.1%) 4 (7.5%) 3 (5.9%) 38 (6.5%) 
Mean (range) 4.7 (2–6) 5.3 (1–24) 6.1 (1–19) 5.0 (1–10) 16.3 (2–25) 6.4 (1–25) 
Gemcitabine N (%) 2 (0.9%) 9 (5.6%) 7 (6.5%) 3 (5.7%) 4 (7.8%) 25 (4.3%) 
Mean (range) 4.5 (2–7) 4.3 (1–13) 3.6 (2–12) 4.3 (2–9) 2.3 (1–4) 3.8 (1–13) 
Epirubicin + IFO N (%) 19 (8.9%) 19 (3.2%) 
Mean (range) 4.9 (1–11)     4.9 (1–11) 
Pazopanib N (%) 8 (5.0%) 2 (1.9%) 1 (1.9%) 5 (9.8%) 16 (2.7%) 
Mean (range)  3.1 (1–14) 6.5 (1–12) 7.0 4.0 (1–7) 4.1 (1–14) 
DTIC + Doxorubicin + IFO N (%) 15 (7.0%) 15 (2.6%) 
Mean (range) 5.6 (4–6)     5.6 (4–6) 
  First line Second line Third line Fourth line Fifth line or later All lines 
All regimens N 213 161 107 53 51 585 
Mean (range) 6 (1–42) 4.9 (1–25) 5.2 (1–24) 4.2 (1–20) 4 (1–25) 5.2 (1–42) 
Doxorubicin N (%) 72 (33.8%) 11 (6.8%) 6 (5.6%) 2 (3.8%) 4 (7.8%) 95 (16.2%) 
Mean (range) 5.6 (2–14) 4.5 (1–6) 4.0 (1–7) 2.0 (2–2) 4.0 (1–10) 5.3 (1–14) 
Trabectedin N (%) 10 (4.7%) 23 (14.3%) 26 (24.3%) 15 (28.3%) 11 (21.6%) 85 (14.5%) 
Mean (range) 14.2 (4–42) 7.1 (1–24) 6.3 (1–23) 5.9 (2–20) 4.5 (1–17) 7.1 (1–42) 
GC + Doc N (%) 23 (10.8%) 29 (18.0%) 10 (9.3%) 5 (9.4%) 67 (11.4%) 
Mean (range) 6.7 (3–14) 5.1 (1–12) 9 (3–24) 4.8 (3–10)  6.2 (1–24) 
Doxorubicin + IFO N (%) 45 (21.1%) 8 (5.0%) 1 (0.9%) 54 (9.2%) 
Mean (range) 5.2 (1–7) 3.6 (2–6) 5.0   5.0 (1–7) 
IFO N (%) 2 (0.9%) 27 (16.8%) 9 (8.4%) 4 (7.5%) 3 (5.9%) 45 (7.7%) 
Mean (range) 3.5 (3–4) 3.9 (1–9) 3.1 (1–6) 3.3 (2–6) 4 (2–6) 3.7 (1–9) 
Investigational N (%) 3 (1.4%) 14 (8.7%) 14 (13.1%) 4 (7.5%) 3 (5.9%) 38 (6.5%) 
Mean (range) 4.7 (2–6) 5.3 (1–24) 6.1 (1–19) 5.0 (1–10) 16.3 (2–25) 6.4 (1–25) 
Gemcitabine N (%) 2 (0.9%) 9 (5.6%) 7 (6.5%) 3 (5.7%) 4 (7.8%) 25 (4.3%) 
Mean (range) 4.5 (2–7) 4.3 (1–13) 3.6 (2–12) 4.3 (2–9) 2.3 (1–4) 3.8 (1–13) 
Epirubicin + IFO N (%) 19 (8.9%) 19 (3.2%) 
Mean (range) 4.9 (1–11)     4.9 (1–11) 
Pazopanib N (%) 8 (5.0%) 2 (1.9%) 1 (1.9%) 5 (9.8%) 16 (2.7%) 
Mean (range)  3.1 (1–14) 6.5 (1–12) 7.0 4.0 (1–7) 4.1 (1–14) 
DTIC + Doxorubicin + IFO N (%) 15 (7.0%) 15 (2.6%) 
Mean (range) 5.6 (4–6)     5.6 (4–6) 

Doc, docetaxel; DTIC, dacarbazine; GC, gemcitabine; IFO, ifosfamide.

Doxorubicin monotherapy was the most common regimen in first line (34%), followed by anthracycline (doxorubicin or epirubicin) plus ifosfamide (30%). In second line, docetaxel plus gemcitabine was the most common regimen (18%). Trabectedin was the most common regimen used in third and later lines. Treatment patterns also varied across countries. Most notably, doxorubicin alone was used less frequently in first line in the United States (1 of 27 patients), Germany (1 of 14), and Italy (none of 17) compared with first line use of doxorubicin alone in The Netherlands (17 of 29), Spain (13 of 25), UK (15 of 30), and France (18 of 60). Supplemental Table S2 and S3 (available at Annals of Oncology online) show first line and second or later lines of chemotherapy by country.

Thirty-nine of 125 patients (31%) treated with doxorubicin reached or exceeded the recommended dose (450 mg/m2). Among the patients exceeding the recommended dose, the median and mean dose were 503 and 526 mg/m2 while the maximum was 994 mg/m2. None of the 16 patients who received epirubicin exceeded the recommended dose. Patients received a variety of investigational agents such as ombrabulin, conatumumab, brostallicin, brivanib, and R1507 (an anti-IGF-1 receptor no longer in development). These drugs were grouped into the category of ‘investigational’.

favorable response to chemotherapy

The number of patients with and without favorable response by cycle is shown in Figure 1. A total of 585 lines of chemotherapy were analyzed, with 314 having favorable response (54%). Favorable response was achieved by 83% of first-line regimens. In second or later lines, the figure ranged from 42% (second line) to 27% (fifth or later line). Favorable response was shown in first-line only in 107 patients (50.2%) and second line or later only in the remaining 106 patents (49.8%).

Figure 1.

Favorable response by line of chemotherapy, all patients.

Figure 1.

Favorable response by line of chemotherapy, all patients.

The distribution of chemotherapy regimens within lines of chemotherapy did not vary for patients with or without favorable response (data not shown). However, unlike the analysis with all lines of therapy, the mean number of cycles of chemotherapy per patient was higher in later lines of therapy in patients with favorable response: first line (mean = 6.6, range 4–42), second line (mean = 7.8, range 4–25), and third line (mean = 8.5, range 2–24). The mean cycle length in lines with a favorable response was 22.7 days.

time spent in health states

Figure 2 shows the mean number of expected weeks per health state. The results highlight the longer time spent off chemotherapy pre-progression following first-line chemotherapy (38 weeks) than second line (5.6 weeks), third line (6.4 weeks), and fourth or later lines (2.7 weeks).

Figure 2.

Mean number of expected weeks per health state.

Figure 2.

Mean number of expected weeks per health state.

reasons for chemotherapy discontinuation

The primary reason for chemotherapy discontinuation in first and second or later lines for those patients who respond favorably is shown in Table 3. In first line, a planned number of cycles given was the most common reason for discontinuation (64%). In second or later lines, disease progression was the most common reason (41%). Anthracyclines alone or in combination were more commonly discontinued due to planned number of cycles given in first line (70% and 71%, respectively) compared with other regimens (20%–57%). Trabectedin was the regimen most frequently associated with discontinuation due to disease progression (60% in first line and 53% second or later lines). Supplemental Table S3 (available at Annals of Oncology online) shows primary reason for discontinuation by country.

Table 3.

Reason for chemotherapy discontinuation in first and second line by mutually exclusive chemotherapy regimen (patients with favorable response within a given line)

First line ATC alone ATC comb GC + Doc ET-743 Other All 
N = 63 N = 72 N = 24 N = 10 N = 7 N = 176 
Planned no. of cycles given 44 (70%) 51 (71%) 11 (46%) 2 (20%) 4 (57%) 112 (64%) 
Toxicity 6 (10%) 7 (10%) 4 (17%) 1 (10%) 18 (10%) 
Disease progression 5 (8%) 3 (4%) 2 (8%) 6 (60%) 1 (14%) 17 (10%) 
Other 4 (6%) 3 (4%) 4 (17%) 11 (6%) 
Maximum benefit obtained 2 (3%) 6 (8%) 1 (10%) 1 (14%) 10 (6%) 
Patient decision 2 (3%) 3 (13%) 1 (14%) 6 (3%) 
Missing 2 (3%) 2 (1%) 
Second or later lines ATC alone ATC comb GC + Doc ET-743 Other All 
N = 11 N = 3 N = 25 N = 34 N = 65 N = 138 
Disease progression 3 (27%) 2 (67%) 11 (44%) 18 (53%) 23 (35%) 57 (41%) 
Planned no. of cycles given 4 (36%) 1 (33%) 8 (32%) 4 (12%) 16 (25%) 33 (24%) 
Other 1 (9%) 1 (4%) 6 (18%) 12 (18%) 20 (14%) 
Toxicity 2 (18%) 3 (12%) 3 (9%) 6 (9%) 14 (10%) 
Maximum benefit obtained 1 (9%) 2 (8%) 1 (3%) 2 (3%) 6 (4%) 
Patient decision 1 (3%) 1 (2%) 2 (1%) 
Missing 1 (3%) 5 (8%) 6 (4%) 
First line ATC alone ATC comb GC + Doc ET-743 Other All 
N = 63 N = 72 N = 24 N = 10 N = 7 N = 176 
Planned no. of cycles given 44 (70%) 51 (71%) 11 (46%) 2 (20%) 4 (57%) 112 (64%) 
Toxicity 6 (10%) 7 (10%) 4 (17%) 1 (10%) 18 (10%) 
Disease progression 5 (8%) 3 (4%) 2 (8%) 6 (60%) 1 (14%) 17 (10%) 
Other 4 (6%) 3 (4%) 4 (17%) 11 (6%) 
Maximum benefit obtained 2 (3%) 6 (8%) 1 (10%) 1 (14%) 10 (6%) 
Patient decision 2 (3%) 3 (13%) 1 (14%) 6 (3%) 
Missing 2 (3%) 2 (1%) 
Second or later lines ATC alone ATC comb GC + Doc ET-743 Other All 
N = 11 N = 3 N = 25 N = 34 N = 65 N = 138 
Disease progression 3 (27%) 2 (67%) 11 (44%) 18 (53%) 23 (35%) 57 (41%) 
Planned no. of cycles given 4 (36%) 1 (33%) 8 (32%) 4 (12%) 16 (25%) 33 (24%) 
Other 1 (9%) 1 (4%) 6 (18%) 12 (18%) 20 (14%) 
Toxicity 2 (18%) 3 (12%) 3 (9%) 6 (9%) 14 (10%) 
Maximum benefit obtained 1 (9%) 2 (8%) 1 (3%) 2 (3%) 6 (4%) 
Patient decision 1 (3%) 1 (2%) 2 (1%) 
Missing 1 (3%) 5 (8%) 6 (4%) 

ATC, anthracycline; ATC comb, anthracycline plus other; Doc, docetaxel; ET-743, trabectedin; GC, gemcitabine.

chemotherapy regimen modification

Chemotherapy was modified in 163 of 585 lines of therapy (28%), most commonly due to intolerable toxicity or adverse events (n = 126, 21%). The percent of patients who modified their chemotherapy regimen was consistent across lines of therapy (range: 23%–32%). Supplemental Table S4 (available at Annals of Oncology online) displays the incidence and reasons for regimen modification per line of treatment.

PFS and OS

Median PFS from first favorable response, across all lines, was 8.3 months [95% confidence interval (CI) 7.4–9.9]. The Kaplan–Meier plot for PFS from first favorable response, stratified by line, is displayed in Figure 3. PFS for patients with first favorable response to first-line treatment was 8.7 months (95% CI 7.9–10.3), 6.6 months (95% CI 2.9–11.2) for second line, and 4.4 months (95% CI 0.8–13.3) for third-line chemotherapy.

Figure 3.

Progression-free survival from date of first favorable response, stratified by line of therapy where favorable response first observed.

Figure 3.

Progression-free survival from date of first favorable response, stratified by line of therapy where favorable response first observed.

Median OS from first favorable response, across all lines, was 23.5 months (95% CI 20.5–28.1). OS from first favorable response, stratified by line, is displayed in Figure 4. Median OS from first favorable response to treatment was 26.1 months (95% CI 20.8–31.4) for first line, 23.8 months (95% CI 12.4–28.1) for second line, and 11.4 months (95% CI 4.1–22.3) for third line, respectively. These differences were statistically significant (P < 0.001). The median OS from diagnosis of metastatic disease was 33.3 months (95% CI 28.0–37.7), 28.9 from initiation of first-line therapy (95% CI 23.6–32.9), and 17.3 months (95% CI 13.1–19.7) from initiation of second line therapy.

Figure 4.

Overall survival from date of first favorable response, stratified by line of therapy where favorable response first observed.

Figure 4.

Overall survival from date of first favorable response, stratified by line of therapy where favorable response first observed.

discussion

Results from this study highlight that PFS and OS in mSTS patients with favorable response to chemotherapy were short. The median PFS and OS from first favorable response to chemotherapy were 8.3 and 23.5 months, respectively. Patients received multiple lines of chemotherapy (average of 2.7, range: 1–8), and the likelihood of favorable response was higher for first line compared with second or later lines of therapy. Distinct chemotherapy treatment patterns emerged for first line compared with second or later lines of therapy. Anthracyclines, with or without ifosfamide, were the dominant regimen in the first-line setting. The common reason for discontinuation of first-line therapy, in lines where a favorable response was observed, was reaching a predefined number of cycles. The average time following first-line chemotherapy when patients were not receiving chemotherapy, but had yet to experience disease progression or start the next line of chemotherapy, was relatively long (38 weeks). In contrast, in second or later lines of therapy, no single dominant regimen emerged and disease progression was the most common reason for discontinuation in lines where a favorable response was observed. In second and later lines, the time period patients were off chemotherapy and prior to disease progression was short (2.7–5.6 weeks). Chemotherapy regimen modification was commonly used across all lines of therapy, with toxicity as the reason in a majority of cases.

The OS from this study can be compared with the control (placebo) arm of the SUCCEED clinical trial, with both studies having similar though not identical entry criteria. The median OS was 19.2 months in the control arm in the SUCCEED trial compared with 23.5 months from first favorable response to chemotherapy in this study [18].

OS statistics with different starting points were determined to compare with observational studies of patients with and without favorable response to chemotherapy [15, 16, 20]. In this study, the median OS was 33.3 months from metastatic disease diagnosis, 28.9 and 17.3 months from initiation of first- and second-line chemotherapy, respectively. Karavasilis et al. [15] estimated 45% of 488 mSTS patients treated at a UK tertiary care institution between 1991 and 2005 derived clinical benefit (stable disease or better for >6 months) from first-line palliative chemotherapy, with a median OS of 12 months from completion of first-line chemotherapy. Using data from the same institution and time period, Minchom et al. [16] reported that 22.7% of 379 patients derive clinical benefit from second-line chemotherapy, with a median OS of 8 months from initiation of second-line chemotherapy. Italiano et al. [20] reported that OS from mSTS diagnosis has improved by 50% during the last 20 years, most likely related to treatment, with a median OS of 18 months (95% CI 15.3–20.7) for 244 patients diagnosed at French Sarcoma Group centers from the most recent period (2002–2006). The longer OS in this study is supported by Reichardt et al. [21], who reported a longer survival for patients with better responses to chemotherapy. Median OS by response type was 10 months for progressive disease, 17 months for stable disease, 25 months for partial response, and was not reached for complete response or no evidence of disease (after a median 33 months follow-up) [21].

Chemotherapy regimens were generally consistent with those recommended in treatment guidelines. Anthracycline regimens, with or without ifosfamide, were the dominant regimen used for first-line therapy accounting for over 70% of utilization, consistent with guidelines where anthracyclines are recommended with the highest level and strength of evidence recommendation [6]. There were no dominant regimens in second or later lines, where lower levels of evidence and consensus for recommended therapies exist compared with first-line therapy [6]. Trabectedin alone was the most commonly used regimen in third through fifth lines of therapy (22%–28%), consistent with its sarcoma indication following failure of anthracyclines and ifosfamide.

From the analysis of time patients spend in each health state, we observed that patients were off chemotherapy in a stable disease or better health state for the longest time period following first-line chemotherapy compared with after second or later lines. This is expected given anthracyclines are most often discontinued when a pre-established number of cycles have been reached, and anthracyclines were used in a majority of first-line regimens. This time period, off cytotoxic chemotherapy and before disease progression, is when metastatic sarcoma patients experience a significantly better health-related quality. A recent cross section survey of 120 patients with metastatic soft tissue or bone sarcoma reported patients in stable disease or better and off chemotherapy had a quality of life score 25 points (out of 100 from the EORTC QLQ C30 overall health status scale, representing quality of life) higher than patients in progressive disease (P < 0.001) and 9, 16, and 20 points higher than patients currently treated with first (P = 0.14), second (P = 0.0092), or third line (P = 0.0054) chemotherapy, respectively [17]. In second and later lines of chemotherapy, patients were more commonly treated to disease progression and their disease remained controlled for a significantly shorter interval before progressing and/or receiving the next treatment.

Recently, maintenance therapy has emerged as a strategy to delay progression of disease in mSTS patients with favorable response (stable disease or better) to chemotherapy [18]. Based on results from this study, the number of patients anticipated to benefit from maintenance therapy in mSTS is the greatest for those with favorable response to first-line chemotherapy, given first-line chemotherapy is more likely to be discontinued after a fixed number of cycles rather than due to disease progression. In addition, there is a higher likelihood of favorable response in the first line than in later lines and less attrition of patients as they progress through lines of therapy.

The major strength of this study is the detailed description of chemotherapy treatment patterns and clinical outcomes in real-life clinical practice across several countries. A study limitation was that approximately one-third of patients were alive at the time of chart abstraction, and some may have gone to receive additional chemotherapies not captured in the descriptive analysis of treatment patterns. Another potential limitation was that the analyses were mostly descriptive, which was necessitated by the sample size. A larger sample may have allowed an evaluation of the association of different treatment patterns with clinical outcomes.

This study highlights the significant unmet medical need in mSTS patients with favorable response to chemotherapy while also providing a description of contemporary treatment across a range of countries. Chemotherapy use is generally consistent with treatment guideline recommendations. The types of chemotherapy used, reasons for discontinuation, and the chemotherapy treatment-free period were quite distinct for first line as compared with second or later lines.

funding

Merck & Co., Inc.

disclosure

PR is a member of Speaker's Bureau and Advisory Board for Merck & Co; AS has conducted research for Ariad Pharmaceuticals; ABB and TB are employees of Merck & Co and potentially own stock and/or hold stock options in Merck; LJ, AM, and NJ have provided consulting services to Merck & Co, Inc; JYB is conducting research sponsored by Merck, as well as GSK, Roche, Novartis, and PharmaMar; ML, XGdM, IJ, and JV have no conflicts to declare.

acknowledgements

We would also like to acknowledge the investigators and study nurses at the study centres: Annette Reichardt, HELIOS Klinikum, Berlin; Sebastian Bauer, Sahra Rauhut, Innere Klinik, Essen; Clemens Wendtner, Lars Pester, Universitätsklinikum, Köln; William Sebag, Hopital Edouard Herriot, Centre Leon Berard, Lyon; Francois Goldwasser, Wilana Jepiral, Hopital Cochin, Paris; Loïc Chaigneau, Ludivine Roy, L'hôpital Jean Minjoz, Besançon Cedex; Binh Bui-Nguyen, Sabrina Albert, Institute Bergonié, Bordeaux Cedex; Umberto Basso, Anna Roma, Istituto Oncologico Veneto, Padova; Virgina Ferraresi, Maria Tedeschi, Istituto Regina Elena, Rome; Emanuela Marchesi, Istituto Ortopedico Rizzoli, Bologna; Jaap Verweij, Marianne Mol, Erasmus Medisch Centrum, Rotterdam; Juliëtte Kamphuis, Leiden University Medical Centre, Leiden; Antonio Casado Herraéz, Hortensia Peleteiro, Hospital Clínico Universitario San Carlo, Madrid; Laura Jiménez, Institut Catalá d'Oncologia, Barcelona; María Gajete, Hospital Son Dureta, Palma De Mallorca; Antonio López Pousa, Rosa Sanchez, Hospital de la Santa Creu i Sant Pau, Barcelona; Hans Hagberg, Uppsala University Hospital, Uppsala; Quincy Chu, Margo Miller, University of Alberta Cross Cancer Institute, Edmonton; Sant P Chawla, Victoria Chu, Sarcoma Oncology Center, Santa Monica; Arthur Staddon, Zackory Burns, Pennsylvania Oncology and Hematology Association, Philadelphia; Brian Samuels, Ryan Dobler, Robin Gustin, Kootenai Cancer Center, Post Falls; Sarah Welby, Christie Hospital, Manchester; Ian Judson, Galina Petrikova, Royal Marsden Hospital, Sutton; Larry Hayward, Paul Ramage, The Queen's Medical Research Institute, Edinburgh.

references

1
Fletcher
CDM
Unni
KK
Mertens
F
World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Soft Tissue and Bone
 , 
2002
Lyon, France
IARC Press
2
Ducimetiere
F
Lurkin
A
Ranchere-Vince
R
, et al.  . 
Incidence of Sarcoma histiotypes and molecular subtypes in a prospective epidemiological study with central pathology review and molecular testing
PLoS One
 , 
2011
, vol. 
6
 
(8)
pg. 
e20294
  
Published on August 3, 2011.
3
Wimber
C
Leithner
A
Zielonke
A
, et al.  . 
Increasing incidence rates of soft tissues sarcoma? A population-based epidemiologic study and literature review
Ann Oncol
 , 
2010
, vol. 
21
 (pg. 
1106
-
1121
)
4
US National Institutes of Health
 
SEER Stat Fact Sheets: Soft Tissue Including Heart. National Cancer Institute, Surveillance Epidemiology and End Results. http//seer.cancer.gov/statfacts/html/soft.html (10 March 2012, date last accessed)
5
Coindre
JM
Terrier
P
Guillou
L
, et al.  . 
Predictive value of grade for metastasis development in the main histologic types of adult soft tissue sarcomas
Cancer
 , 
2001
, vol. 
91
 (pg. 
1914
-
1926
)
6
Casali
PG
Blay
JY
Soft tissue sarcomas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up
Ann Oncol
 , 
2010
, vol. 
21
 
(Suppl 5)
(pg. 
v198
-
v203
)
7
NCCN Clinical Practice Guidelines in Oncology
Soft Tissue Sarcoma, version 1
 , 
2011
 
8
Grimer
R
Judson
I
Peake
D
Seddon
B
Guidelines for the management of soft tissue sarcomas
Sarcoma
 , 
2010
 
Accepted February 3, 2010. Article ID 506182,
9
Krikelis
D
Judson
I
Role of chemotherapy in the management of soft tissue sarcomas
Expert Rev Anticancer Ther
 , 
2010
, vol. 
10
 
2
(pg. 
249
-
260
)
10
Maki
RG
Wathen
JK
Patel
SR
, et al.  . 
Randomized phase II study of gemcitabine and docetaxel compared with gemcitabine alone in patients with metastatic soft tissue sarcomas: results of sarcoma alliance for research through collaboration study 002
J Clin Oncol
 , 
2007
, vol. 
25
 
19
(pg. 
2755
-
2763
)
11
Duffaud
F
Bui
BN
Penel
N
, et al.  . 
A FNCLCC French Sarcoma Group—GETO multicenter randomized phase II study of gemcitabine (G) versus gemcitabine and docetaxel (G+D) in patients with metastatic or relapsed leiomyosarcoma (LMS)
J Clin Oncol
 , 
2008
, vol. 
26
 
(Suppl)
 
(Abstr 10511)
12
Schlemmer
M
Reichardt
P
Verweij
J
, et al.  . 
Paclitaxel in patients with advanced angiosarcomas of soft tissue: a retrospective study of the EORTC soft tissue and bone sarcoma group
Eur J Cancer
 , 
2008
, vol. 
44
 (pg. 
2433
-
2436
)
13
Demetri
GD
Chawla
SP
von Mehren
M
, et al.  . 
Efficacy and safety of trabectedin in patients with advanced or metastatic liposarcoma or leiomyosarcoma after failure of prior anthracyclines and ifosfamide: results of a randomized phase II study of two different schedules
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
4188
-
4196
)
14
Gross
F
Sanfillippo
R
Virdis
E
, et al.  . 
Trabectedin in myxoid liposarcomas (MLS): a long-term analysis of a single-institution series
Ann Oncol
 , 
2009
, vol. 
20
 
8
(pg. 
1439
-
1444
)
15
Karavasilis
V
Seddon
BM
Ashley
S
, et al.  . 
Significant clinical benefit of first-line palliative chemotherapy in advanced soft-tissue sarcoma
Cancer
 , 
2008
, vol. 
112
 (pg. 
1585
-
1590
)
16
Minchom
A
Jones
RL
Fisher
C
, et al.  . 
Clinical benefit of second-line palliative chemotherapy in advanced soft-tissue sarcoma
Sarcoma
 , 
2010
 
Published May 20, 2010. Article ID 264360,
17
Reichardt
P
Leahy
M
Garcia del Muro
X
, et al.  . 
Quality of life and utility in patients with metastatic soft tissue and bone sarcoma: the sarcoma treatment and burden of illness in North America and Europe (SABINE) study
Sarcoma
 , 
2012
 
in press
18
Chawla
SP
Blay
JY
Ray-Coquard I et al. Results of the phase III, placebo-controlled trial (SUCCEED) evaluating the mTOR inhibitor ridaforolimus (R) as maintenance therapy in advanced sarcoma patients (pts) following clinical benefit from prior standard cytotoxic chemotherapy (CT)
J Clin Oncol
 , 
2011
, vol. 
29
 
(Suppl)
 
(Abstr 10005)
19
Product Information: Epirubicin Hydrochloride IV Injection, Epirubicin Hydrochloride IV Injection
 , 
2010
Irvine, CA
Teva Parenteral Medicines, Inc.
20
Italiano
A
Mathoulin-Pelissier
S
Cesne
AL
, et al.  . 
Trends in survival for patients with metastatic soft-tissue sarcoma
Cancer
 , 
2011
, vol. 
117
 (pg. 
1049
-
1054
)
21
Reichardt
P
Tilgner
J
Hohenberger
P
Dorken
P
Dose-intensive chemotherapy with ifosfamide, epirubicin, and filgrastim for adult patients with metastatic or locally advanced soft tissue sarcoma: a phase II study
J Clin Oncol
 , 
1998
, vol. 
16
 (pg. 
1438
-
1443
)