The real‐world clinical outcomes and treatment patterns of patients with unresectable locally advanced or metastatic soft tissue sarcoma treated with anlotinib in the post‐ALTER0203 trial era

Abstract Background The ALTER0203 clinical trial showed that anlotinib, a multitargeted tyrosine kinase inhibitor, had antitumor effects on advanced soft tissue sarcoma (STS) after the failure of standard chemotherapy. We aimed to evaluate the real‐world efficacy and explore prognostic factors and treatment patterns of anlotinib in patients with advanced STS. Methods We retrospectively analyzed the data of patients with unresectable locally advanced or metastatic STS who received at least one dose of anlotinib from June 2018 to March 2021. The survival data were analyzed using the Kaplan–Meier method and compared using the log‐rank test. The Cox proportional hazards model was performed for multivariate analysis. Results A total of 209 patients were included. The median age was 48 (range 11–85) years. The median follow‐up, progression‐free survival, and overall survival were 18.7 months, 6.1 months [95% confidence interval (CI): 4.9–7.2], and 16.4 months (95% CI: 13.6–19.1), respectively. The objective response rate was 13.4%. Nutritional status, Eastern Cooperative Oncology Group (ECOG) performance status, and anlotinib treatment patterns (combination therapy or switch maintenance therapy vs. monotherapy) were significantly associated with progression‐free survival. Besides, pathological grade, nutritional status, ECOG performance status, and anlotinib treatment patterns were predictive of overall survival. Due to anlotinib‐related toxicity, 31 (14.8%) patients, and 25 (12.0%) patients experienced dose reduction and treatment discontinuation, respectively. Conclusion These findings confirmed the efficacy of anlotinib in patients with advanced STS in a real‐world setting. The patterns of anlotinib treatment deserve further exploration.


| INTRODUCTION
Soft tissue sarcomas (STS) are a heterogeneous group of rare tumors that account for 0.8-1% of all adult malignancies. STS displays different clinical manifestations and clinical courses with more than 50 different pathological subtypes. 1 The prognosis for patients with unresectable locally advanced or metastatic STS remains poor. The cornerstone therapy of advanced STS therapy is anthracycline-based chemotherapy which has a median overall survival of approximately 1 year. [2][3][4][5] In the past decade, advances in the treatment of unresectable or metastatic STS have included novel chemotherapy agents such as trabectedin and eribulin, 6,7 highly selective tyrosine kinase inhibitor (TKI) such as crizotinib and imatinib, 8,9 and multitargeted antiangiogenic TKI such as pazopanib and regorafenib. 10,11 Another multitargeted TKI, anlotinib exerts inhibitory effects against vascular endothelial growth factor receptors, platelet-derived growth factor receptors, and fibroblast growth factor receptors. 12 The phase IIb trial (ALTER0203) was conducted in the Chinese population to explore the antitumor activity of anlotinib in standard chemotherapy-failed advanced STS, including leiomyosarcoma (LMS), synovial sarcoma (SS), alveolar soft part sarcoma (ASPS), clear cell sarcoma, epithelioid sarcoma, undifferentiated pleomorphic sarcoma, liposarcoma, and fibrosarcoma. 13 The results showed that the median progression-free survival (PFS) of the anlotinib group was significantly longer than that of the placebo group (6.27 vs. 1.47 months), and patients with LMS, SS, or ASPS subtypes were more likely to benefit from anlotinib treatment. 13 Based on these data, anlotinib has been approved in China for the treatment of advanced STS. The international phase III randomized clinical trial (APROMISS) comparing the efficacy of anlotinib and dacarbazine in patients with SS showed that anlotinib improved median PFS (2.89 vs. 1.64 months). 14 However, both clinical trials strictly selected age, Eastern Cooperative Oncology Group (ECOG) performance status, and pathological subtypes.
Except for pathological subtypes, a study showed that anlotinib treatment-related adverse events, including increased thyroid stimulating hormone, hypertriglyceridemia, and proteinuria were associated with longer PFS in patients enrolled in the ALTER0203 study. 15 Patients with longer PFS may have prolonged exposure to drug therapy and may be at higher risk for treatment-related adverse events. Therefore, the relationship between PFS and treatment-related adverse events seems to be indistinct.
In this study, we evaluated the real-world efficacy of anlotinib in patients with advanced STS and explored prognostic factors and treatment patterns in the post-ALTER0203 trial era.

| Study design
This study was a retrospective study approved by the Ethics Committee of Biomedical Research, West China Hospital of Sichuan University. The requirement for individual informed consent was waived. Data were obtained from the Soft Tissue Sarcoma Database of Cancer Center, West China Hospital.
This study included patients who were pathologically diagnosed with STS, had unresectable locally advanced or metastatic STS lesions, and received at least one dose of anlotinib from June 2018 to March 2021. Malignant phyllodes tumor of breast with sarcoma component and dedifferentiated chordoma were also allowed. There were no patients excluded.
The following baseline clinicopathologic data were recorded: age, gender, pathological subtypes, pathological grade according to the French Federation of Cancer Centers Sarcoma Group (FNCLCC) systems, 16 primary and metastatic sites, largest or symptomatic lesion, tumor size, number of lesions, pretreatment body mass index (BMI), weight change within 6 months before anlotinib therapy, serum albumin and lactate dehydrogenase level, ECOG performance status, comorbidity status, and treatment history. BMI was calculated by body weight (kg) divided by the square of height (m 2 ). According to the consensus for cachexia, 17 malnutrition in the present study was defined as body weight loss >5% in 6 months, weight loss >2% and BMI <20 kg/m 2 , or serum albumin level below the normal lower limit. Switch maintenance therapy referred to patients who achieved disease control after previous chemotherapy were given a sequential treatment of anlotinib until disease progression or unacceptable toxicity. Combination therapy was defined as patients who received other antitumor treatments during anlotinib treatment, including surgery, radiofrequency ablation, radiation, chemotherapy, or immunotherapy.

K E Y W O R D S
clinical management, nutrition, prognosis, soft tissue sarcoma, target therapy, tyrosine kinase inhibitors

| Treatment
All patients received a once daily oral anlotinib at a dose of 12, 10 mg, or 8 mg based on the patients' conditions. The treatment was administered for 1-14 days in 21-day cycles. The drug dose of anlotinib was reduced for patients who experienced intolerable adverse events (AEs). Patients continued anlotinib treatment until intolerable AEs or progressive disease (PD) occurred. AEs were assessed using the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0.

| Treatment assessment
Treatment response was assessed according to RECIST version 1.1 criteria. 18 The objective response rate (ORR) was the sum of the rates of confirmed complete response (CR) and partial response (PR). Stable disease should last at least 6 weeks. PFS was calculated from the initiation of anlotinib treatment to disease progression or death from any cause, and overall survival (OS) was calculated from the initiation of anlotinib treatment to death from any cause.

| Statistical analysis
Descriptive statistics were used to summarize patient characteristics. For patients lost to follow-up, the last time of follow-up was considered the time of death. The cutoff date was August 25, 2021. The Kaplan-Meier method was used to represent the survival curves of PFS and OS, and the log-rank test was used for comparison. Factors that were statistically associated with survival according to the log-rank tests were included in the collinearity analysis. Variables with weak collinearity were included in the multivariate Cox proportional regression hazards models to explore independent prognostic factors. A twosided p < 0.05 was considered statistically significant. The results were described as hazard ratio (HR) with a 95% confidence interval (CI). The statistical software was SPSS (version 17.0), and GraphPad Prism 8.0 was used for plotting.
The most common primary tumor site was the trunk and extremities (43.1%). The majority of patients (83.7%) had one or more metastases, and the most common site was the lung (54.1%). The largest group of pathological subtypes was leiomyosarcoma (18.7%). According to the FNCLCC grading system, 153 (73.2%) cases were grade 2-3. Thirty-nine (18.7%) patients had an ECOG performance status ≥2. Forty (19.1%) patients had lactate dehydrogenase above the upper limit of normal, and 25 (12.0%) patients with unknown lactate dehydrogenase were classified as normal group. The baseline characteristics are outlined in Table 1.

| Nutritional assessment
The mean (±SD) BMI was 22.4 ± 3.5 kg/m 2 . Fifty-three (25.4%) patients had a BMI ≤20 kg/m 2 . The BMI of five patients was unknown and was replaced with the mean value. Seventy-eight (37.3%) patients lost weight before anlotinib treatment, of whom 52 (24.9%) lost more than 5% of their body weight. Unknown weight change in eight (3.8%) patients was considered as no weight change. Fiftytwo (24.9%) patients had serum albumin below the lower limit of normal, and 26 (12.4%) patients with unknown albumin were classified as normal group. A total of 85 (40.7%) patients met the criteria for malnutrition.

| DISCUSSION
To the best of our knowledge, this study represented the largest sample size focusing on the clinical outcomes of patients with unresectable locally advanced or metastatic  STS treated with anlotinib in the post-ALTER0203 trial era. The median PFS, median OS, and ORR were 6.1 months, 16.4 months, and 13.4%, respectively (Figure 1). The rates of dose reduction and treatment discontinuation caused by adverse events were 14.8% and 12.0%, respectively. Multitargeted antiangiogenic therapy is an important treatment in patients with advanced non-adipocytic STS after the failure of standard chemotherapy. In this population, the median PFS with pazopanib and regorafenib treatment was 4.6 and 4.0 months, respectively. 10,11 For patients treated with anlotinib, the median PFS was 6.27 months in the ALTER0203 study, which included eight pathological subtypes, 13 and 2.89 months in the APROMISS study, which focused on one pathological subtype, synovial sarcoma. 14 Our study included approximately 30 pathological subtypes without selection and exclusion criteria. The median PFS and ORR were 6.1 months and 13.4% for all patients and 4.0 months and 8.0% for patients treated with anlotinib monotherapy, respectively. Only 14.8% of patients had dose reduction, and 12.0% discontinued anlotinib treatment due to adverse events. Our study showed that anlotinib was effective and tolerable in patients with advanced STS in real-world clinical practice, similar to the above studies. 13,14 F I G U R E 2 Survival curves of progression-free survival according to nutritional status (A), ECOG performance status (B), and anlotinib treatment patterns (C). Survival curves of overall survival according to nutritional status (D), ECOG performance status (E), and anlotinib treatment patterns (F) In the ALTER0203 clinical trial, the median PFS with anlotinib treatment was longer in females and patients aged >40 years. 20 Our study found that neither age nor gender was associated with PFS. One possible reason is the differences in the range of age (18-70 vs. 11-85 years), ECOG performance status (0-1 vs. 0-3), number of pathological subtypes (8 vs. approximately 30), and treatment patterns (monotherapy vs. three patterns) between the ALTER0203 study and our study. However, similar to the finding of the ALTER0203 study that LMS, SS, and ASPS subtypes were more sensitive to anlotinib treatment, our results showed that these three subtypes were associated with longer OS according to univariate analysis (Table 4) but not an independent prognostic factor according to multivariate analysis (Table 5). However, the results of multivariate analysis provided evidence of the prognostic significance of FNCLCC grades in our study (Table 5), which is one of the most commonly used pathological grading systems for STS. 16 Malnutrition is a severe problem affecting prognosis and quality of life in patients receiving anticancer therapy. Several retrospective studies reported that BMI and weight loss were associated with survival in patients with solid tumors who received TKI therapies. [21][22][23][24] However, changes in body weight may be masked by weight gain in ascites, peripheral edema, and growth of primary and metastatic tumors. 25,26 Therefore, in addition to sustained weight loss >5% or weight loss >2% with BMI <20 kg/m 2 , sarcopenia was also taken into consideration in the international consensus for cachexia diagnostic criteria. 17 Measurement of sarcopenia requires an assessment of muscle mass and strength, of which direct measurement, including crosssectional imaging, is preferred. However, there was no clear consensus on screening tools for sarcopenia, and testing methods were complex and inconvenient, making universal measurement difficult for patients. 19 Several studies have shown that low serum albumin levels are an independent poor prognostic factor in patients with STS and are associated with poor survival. [27][28][29][30] The serum albumin test is convenient and repeatable. In the present study, a diagnosis of malnutrition was made if patients met one of the following criteria: body weight loss >5% in 6 months, weight loss >2% and BMI <20 kg/m 2 , or serum albumin level below the normal lower limit. These malnutrition criteria might have prognostic power in advanced STS patients treated with anlotinib, as demonstrated by the significant association of malnutrition with survival outcomes (Tables 4 and 5, Figure 2).
The ECOG performance status reflecting the general condition of patients is a common inclusion criterion in clinical trials. Although cancer patients with poor performance status may still benefit from antitumor treatment, 31 most cancer clinical trials had strict performance status (0 or 1) criteria, 32 such as the PALETTE, REGOSARC, and ALTER0203 studies. 10,11,13 Indeed, we found that patients (n = 39, 18.7%) with an ECOG performance status of ≥2 had poor survival outcomes (Tables 4 and 5, Figure 2), which was consistent with the findings in patients with advanced non-small cell lung cancer receiving anlotinib treatment. 33,34 From the data in Table 1, it was observed that ECOG performance status also generally determined the treatment strategy in clinical practice. Therefore, we conducted an ad hoc analysis to explore the survival difference in treatment patterns stratified by ECOG performance status. Among patients with a performance status of ≥2, there was no significant difference in survival outcomes with combination therapy or anlotinib monotherapy, but in the subset of patients with a performance status of 0 or 1, both anlotinib combination therapy and switch maintenance therapy provided improvements in median PFS and OS compared with anlotinib monotherapy. Although multitargeted TKI has been an important second-line treatment option for patients with advanced STS after the failure of standard chemotherapy, the strategies of targeted treatment require further studies. Combination with antiangiogenic TKI and immune checkpoint inhibitor was a promising strategy, 35,36 but faced the selection of pathological subtype. Another strategy is antiangiogenic TKI in combination with chemotherapy. For 27 patients treated with anlotinib combined with liposomal doxorubicin followed by anlotinib maintenance, the ORR and median PFS were 40.7% and 7 months, respectively. 37 A recent phase II clinical study reported that the ORR and median PFS were 13.3% and 11.5 months, respectively, in patients treated with epirubicin combined with anlotinib followed by anlotinib maintenance. 38 Neither study reported OS data due to the duration of follow-up. The PAPAGEMO phase II trial was conducted in patients with anthracycline-and/or ifosfamide-failed STS and showed that pazopanib combined with gemcitabine significantly prolonged the median PFS compared with pazopanib monotherapy, but with similar median OS. 39 Our findings suggested that compared with anlotinib monotherapy, anlotinib in combination with other therapies not only improved the ORR and median PFS, but also prolonged the median OS (Tables 4 and 5, Figure 2). In addition to combining chemotherapy, anlotinib was also combined with immunotherapy or local treatment in some cases. This may explain the difference in the findings of the PAPAGEMO study 39 and our study. Promisingly, switching maintenance therapy with anlotinib after chemotherapy was also significantly associated with longer median PFS and OS (Tables 4 and 5, Figure 2). Possible reasons were that achieving an objective response or stable disease after chemotherapy may have selected patients with good prognosis, and the delayed effects of chemotherapy may add to anlotinib maintenance therapy. 40 Many trials investigating the strategies of anlotinib treatment in patients with locally advanced or metastatic STS are currently active, such as anlotinib in combination with chemotherapy (NCT03880695, NCT03815474), anlotinib in combination with immune checkpoint inhibitors (NCT04172805, NCT04165330), and switch maintenance therapy with anlotinib after chemotherapy (NCT03890068). However, prospective randomized controlled trials are still lacking.
Several limitations existed in this study. First, there might be selection bias in our data due to the nature of the single-institutional retrospective study design and the heterogeneity of STS. To reduce selection bias as far as possible, we continuously collected cases from June 2018 to March 2021, and we did not select patients' age, pathological subtypes, or ECOG performance status. Second, the generalizability of our study is unclear because patients in our center mainly came from southwest China. However, the population of Southwest China was over 200 million in 2020. Our center, as a department of a large universityaffiliated tertiary comprehensive hospital, has extensive experience in the management of advanced STS in this region. Third, the sample size of 209 was not very large but had preciousness due to the rarity of STS.

| CONCLUSIONS
Our findings confirmed the antitumor activity of anlotinib in patients with advanced STS in a real-world setting. The FNCLCC grade, nutritional status, ECOG performance status, and treatment patterns were predictive of survival in this population. The appropriate treatment patterns for anlotinib deserve further exploration.
University. The requirement for individual informed consent was waived.

DATA AVAILABILITY STATEMENT
All data are available from the corresponding author.