Comparison of clinical features and outcomes of patients with leiomyosarcoma of bone and soft tissue: a population-based cohort study

Abstract

Background

Leiomyosarcoma commonly occurs in soft tissue but rarely in the bone. Whether leiomyosarcoma of bone and soft tissue have similar clinical characteristics and outcomes remains unknown.

Methods

This retrospective analysis was based on data from the Bone and Soft Tissue Tumor Registry in Japan. Patients with leiomyosarcoma of bone and soft tissue were enrolled. Overall survival and distant metastasis-free survival were estimated using the Kaplan–Meier method, and the Cox regression model was used to identify the prognostic factors.

Results

A total of 888 patients (60 leiomyosarcoma of bone and 828 leiomyosarcoma of soft tissue) were included in the study. Clinical characteristics were similar between the two groups, except for younger age in leiomyosarcoma of bone than in leiomyosarcoma of soft tissue (median 56 years vs. 66 years, P < 0.0001). To evaluate the prognostic factors and efficacy of adjuvant chemotherapy, data of localized and locally curative cases were extracted (total 572: 33 leiomyosarcoma of bone and 539 leiomyosarcoma of soft tissue). The 5-year overall survival rates of leiomyosarcoma of bone and soft tissue patients were similar (63.8% vs. 75.2%, P = 0.43); the 5-year distant metastasis-free survival tended to be worse in leiomyosarcoma of bone than in leiomyosarcoma of soft tissue (37.4% vs. 57.9%, P = 0.28). Larger tumor size (≥5 cm) and older age (≥65 years) correlated with poor overall survival in leiomyosarcoma of soft tissue patients. Adjuvant chemotherapy tended to prolong the overall survival of both leiomyosarcoma of bone (P = 0.11) and leiomyosarcoma of soft tissue patients with tumor size >10 cm (P = 0.06).

Conclusions

The clinical characteristics and outcomes of leiomyosarcoma of bone and soft tissue patients were similar. In localized cases, adjuvant chemotherapy may improve the survival of leiomyosarcoma of bone and soft tissue patients with large-size tumor.

Introduction

Leiomyosarcoma (LMS) is a malignant neoplasm diagnosed based on tumor cells with smooth muscle differentiation, which accounts for ~11% of all newly diagnosed soft tissue sarcomas (STSs) (1). Although most LMSs arise in the extremities, uterus and retroperitoneum, they may occur in other viscera, including the bladder and gastrointestinal tract (2–5).

The impact of primary site on LMS behavior and prognosis remains controversial. Based on several studies investigating the clinicopathologic differences of LMS tumors with different sites of origin, including the uterus, extremities, trunk, bladder and retroperitoneum, LMS tumors in different sites are not thought to be biologically distinct (3–5). However, a recent molecular and genomic study evaluating potential differences in LMS according to the site of origin demonstrated that LMS could be classified based on the distinct lineages of smooth muscle cells (6).

Sarcomas may arise from both bone and soft tissue. For example, the osteosarcoma of skeletal and extraskeletal origin has distinct clinical features and outcomes, and the efficacy of perioperative chemotherapy differs with the site of origin (7). Similarly, Ewing sarcoma of skeletal and extraskeletal origin has different clinical features, regardless of their biological differences, which remain unknown (8–10).

There are several reports on LMS that have investigated LMS of bone (LMS-B) and LMS of soft tissue (LMS-ST) independently (11–15); however, only one report has evaluated and compared the clinical characteristics, outcomes and prognostic factors of LMS-B and LMS-ST (16). The study investigated various cases, especially those with unresectable primary lesions and distant metastases at first presentation, using the Surveillance, Epidemiology and End Results database. Data on LMS-B are limited due to the disease’s rarity. Moreover, it has not been elucidated whether LMS-B should be treated like LMS-ST or other bone sarcomas. Therefore, the differences in the clinical characteristics and clinical outcomes between LMS-B and LMS-ST are needed to know the similarity or difference between the two LMS variants and the appropriate treatment strategy.

In this study, we aimed to assess the clinicopathologic features, recurrence patterns and survival outcomes of patients with LMS-B and LMS-ST. Moreover, we identified the prognostic factors and especially, the impact of adjuvant chemotherapy on survival rates in patients with localized and resectable primary lesions, using the Japanese Nationwide Bone and Soft Tissue Tumor Registry.

Patients and methods

Data source

This study was conducted using the Japanese Nationwide Bone and Soft Tissue Tumor Registry. The registry was launched by the Japanese Orthopaedic Association (JOA), with an objective to collect data for bone and soft tissue tumors from JOA-certified hospitals for musculoskeletal oncology. The data categories of the registry include patients’ demographic characteristics, clinical and pathological features of the tumor, details of therapeutic intervention and prognostic outcomes at the time of the latest follow-up. Details of the database have been described elsewhere (17). Clinical research using the registry was approved by the Musculoskeletal Tumor Committee of the JOA in 2014 (18), and approval for this study was obtained from the institutional review board (IRB) of the JOA. Because the data extracted from the database are de-identified, the requirement of informed consent was waived by the IRB.

Patient selection and data extraction

We searched the Bone and Soft Tissue Tumor Registry, Japan, for the period between 2004 and 2019 and included all patients diagnosed with LMS-B or LMS-ST with sufficient available prognostic information. For each patient, we extracted the following data: patient age, sex, tumor size, histological grade, the primary location of the tumor, status at first visit and therapeutic intervention (surgical and non-surgical treatments). Age was categorized as follows: 0–39 years [children and adolescents/young adults (AYAs)], 40–64 years (adults) and ≥ 65 years (elderlies). Tumor locations were grouped into the upper and lower extremities and trunk. Tumor size was classified in accordance with the AJCC 8th Edition Staging System; however, soft tissue tumors that were above 15 cm were included in the group of tumors above 10 cm because of the small sample size. Histological grade was classified into low grade or high grade.

Distant metastasis-free survival (DMFS) and overall survival (OS) were defined as the time from diagnosis to detection of metastasis and death due to any cause, respectively. Regarding the survival analysis, individuals who were alive at the final follow-up were censored at that time point. Cases with insufficient data were excluded from this study.

Statistical analysis

Fisher’s exact test and Wilcoxon test were used to identify differences between categorical and continuous variables, respectively. DMFS and OS were estimated using the Kaplan–Meier method, and cumulative survival rates were compared using the log-rank test, with P < 0.05 considered significant. We performed univariate and multivariate analyses using a Cox proportional hazards model to explore the potential prognostic factors. The results of the analyses were reported as adjusted hazard ratios with 95% confidence intervals and respective P values (two-sided alpha =0.05). All statistical analyses were performed using JMP® 13 (SAS Institute Inc., Cary, NC, USA).

Results

Clinical features and treatment strategies of LMS-B and LMS-ST

In total, 1888 cases indicating LMS were identified, and data of 888 cases were extracted by the inclusion criteria as mentioned before. Of these patients, 828 (93.2%) presented with LMS-ST, and only 60 (6.8%) had LMS-B. A comparison of the clinical characteristics of patients with LMS-B and LMS-ST is shown in Table 1. Regarding age at diagnosis, patients with LMS-ST were older than those with LMS-B (median age, 66 years vs. 58 years; P < 0.0001). There was no difference in the primary location and primary tumor size between the two groups. The rate of bone metastasis at first presentation was slightly higher in patients with LMS-B than in patients with LMS-ST (5% vs. 1.9%; P = 0.17), and the rate of lymph node metastasis at first presentation was recognized only in LMS-ST cases. Both groups included a few cases of low-grade tumors (1.7% for LMS-B vs. 7.3% for LMS-ST).

Regarding treatment strategies, patients with LMS-B were more likely to receive adjuvant chemotherapy than those with LMS-ST (53.3% vs. 20.1%, P < 0.0001). Regarding local therapy, patients with LMS-B were less likely to undergo surgery for the primary lesion than those with LMS-ST (87.7% vs. 71.7%, P = 0.002). More patients with LMS-B tended to undergo amputation than those with LMS-ST (9.3% vs. 4.8%, P < 0.0001), but the surgical margins evaluated postoperatively were similar between the two groups.

The DMFS and OS rates are shown in Fig. 1. DMFS of LMS-B tended to be worse than that of LMS-ST, and the 5-year DMFS rates were 32.7 and 51.1% in patients with LMS-B and LMS-ST, respectively; however, the difference was not statistically significant (P = 0.28). The rate of distant metastasis during the time course was similar for patients with LMS-B and LMS-ST (51.7% vs. 44.6%, P = 0.33); however, patients with LMS-ST were more likely to experience lymph node metastasis (0% vs. 3.3%, P = 0.05). The rate of bone metastasis was significantly higher in patients with LMS-B than in those with LMS-ST (18.3% vs. 5.3%, P = 0.0007). OS was similar for both groups (P = 0.43), and the 5-year survival rate was 57.2 and 66.4% in patients with LMS-B and LMS-ST, respectively.

Prognostic factors and impact of adjuvant chemotherapy of localized high-grade LMS-B and LMS-ST1

To compare clinical outcomes of LMS-B and LMS-ST, especially the effect of adjuvant chemotherapy, patient data were extracted according to the following criteria: (i) localized cases, that is, no metastasis at first presentation; (ii) high grade; and (iii) treatment with radical local therapy, resection or amputation. We excluded patients with a primary tumor in the retroperitoneum, peritoneum, thoracic cavity, mediastinum, vertebra, head and neck, because local wide excision with curative intent is relatively difficult because of anatomical sites.

Table 1 shows the clinical characteristics of the extracted cases (N = 572; LMS-B 33, LMS-ST 539). Data of the extracted cases showed similar clinical characteristics in all cases; patients with LMS-B were younger and received adjuvant chemotherapy and amputation more frequently than those with LMS-ST.

The OS and DMFS estimates are shown in Fig. 2. The 5-year OS and DMFS rates were lower in LMS-B cases, with 57.9 and 37.4% in patients with LMS-B, respectively; and 75.2 and 63.8% in patients with LMS-ST, respectively, but the differences were not statistically significant between the two groups (P = 0.89 and P = 0.37, respectively). For patients with LMS-ST experiencing distant metastasis, surgery for metastasis significantly improved OS compared with no surgery for metastasis (P < 0.0001). Only three (9.1%) patients with LMS-B experienced local recurrence (at 3, 15 and 46 months), and 47 (8.7%) patients with LMS-ST experienced local recurrence at a median of 14 months (interquartile range 5–22 months).

The univariate analysis of OS revealed that no factors were related to OS in patients with LMS-B (Table 2). Older age (≥65 years compared with <40 years; P = 0.01) and larger tumor (size ≥ 5 cm, <10 cm; P < 0.0001, size ≥ 10 cm; P < 0.0001) were poor prognostic factors for LMS-ST (Table 2); this was similar in the multivariate analysis (Table 3). Regarding DMFS, the same prognostic factors, older age and larger tumor size, were identified only in patients with LMS-ST by univariate and multivariate analyses (Tables 2 and 3).

A total of 22 (66.7%) patients with LMS-B and 118 (21.9%) patients with LMS-ST received adjuvant chemotherapy (Table 4). Patients with a younger age and larger tumor size were more likely to receive chemotherapy in both the LMS-B and LMS-ST groups. The rates of patients who received chemotherapy with respect to age are shown in Fig. 3. For LMS-B, the rate of patients treated with adjuvant chemotherapy gradually decreased with increasing age (Fig. 3A). For LMS-ST, patients aged <40 years and those aged ≥40 or < 65 years received chemotherapy at comparable rates, with 33.3 and 38.5%, respectively. Patients aged ≥65 years were less likely to receive adjuvant chemotherapy (11.5%, P < 0.0001; Fig. 3B). Patients with LMS-B who received adjuvant chemotherapy tended to have better OS than those who did not (5-year OS: 80.9% vs. 33.3%, P = 0.11, Fig. 4). For patients with LMS-ST, adjuvant chemotherapy did not affect OS in the entire cohort; however, patients with tumor size ≥10 cm treated with adjuvant chemotherapy tended to have better OS than those without adjuvant chemotherapy (5-year OS: 67.9% vs. 53.6%, P = 0.06, Fig. 4).

Discussion

The present study compared the clinical characteristics and outcomes of patients with LMS-B and LMS-ST and investigated the efficacy of adjuvant chemotherapy in both groups, limiting the analysis to cases with localized high-grade tumors in which anatomical sites were appropriate for curative resection. The results showed that (i) the clinical characteristics and outcomes were similar between the two groups, including tumor size, grade and location, except for the age at first presentation and extrapulmonary metastatic sites, and (ii) adjuvant chemotherapy may improve the survival of patients with LMS-B and LMS-ST with large-size tumor.

The clinical characteristics and outcomes of patients with LMS-B and LMS-ST were similar, although several differences were revealed. Similar to the observations in other sarcomas that arise from bone and soft tissue, patients with LMS-B were younger than those with LMS-ST. For example, skeletal osteosarcoma, which is the most common malignant bone tumor, affects children and AYA of ~20–30 years of age, whereas extraskeletal osteosarcoma affects the elderly of ~60 years of age (19). Similarly, extraskeletal Ewing sarcoma occurs in older patients compared with Ewing sarcoma in the bone (8–10). Another difference between bone and soft tissue lesions was the site of distant metastasis. The rate of pulmonary metastasis was similar in both groups; however, LMS-B metastasized to the bone more frequently than LMS-ST at the first presentation and during the treatment course, whereas LMS-ST metastasized to the lymph node. These results are consistent with previous findings: bone sarcoma usually tends to metastasize to bones and STS to the lymph node as extrapulmonary metastasis (10,19,20). Our results indicate that the location of the tumor, rather than the nature of LMS, determines the preferred metastatic site. This evidence indicates that more attention to the extrapulmonary metastatic sites should be paid, depending on the primary site at the first presentation and during follow-up after the primary treatment. Previous studies comparing LMS of the uterus, extremities, retroperitoneum and bladder revealed similar clinical outcomes (3–5); however, a recent genomic study comparing the normal skeletal, vascular, digestive organs and uterine smooth muscle cells revealed that LMS could be classified into three types that have different genomic landscapes and that extremity LMSs were classified into ‘dedifferentiated’ or ‘abdominal/extremity’ subtypes, which resembled the gene expression of vascular smooth muscles (6). Combining the findings of this basic research with our results: LMS-B and LMS-ST showed similar clinical manifestations, it can be interpreted that the cells of origin of LMS-B and LMS-ST are basically the same: somatic vascular smooth muscle cells in the soft tissue and bone. However, no studies have been conducted to compare the genomic changes directly between LMS-B and LMS-ST, which is an issue that needs to be addressed in the future.

The second finding was that adjuvant chemotherapy tended to prolong the OS and DMFS in patients with LMS-B and LMS-ST with large-size tumor. There are several reports about the prognostic factors of LMS, including LMS-B and LMS-ST, which investigated LMSs of various origins, histological grades and primary sites of the tumor (11–16). These reports identified distant metastasis at first presentation, high-grade tumor, inadequate surgical margin and larger tumor size as poor prognostic factors, whereas adjuvant chemotherapy was not a prognostic factor. However, resectability of the primary lesion with curative intent depends on the location of the primary tumor and histological grade influences the efficacy of chemotherapy. Therefore, we analyzed only localized high-grade cases, including those with a primary tumor at the extremity, pelvis and body wall. Through this analysis, we revealed that adjuvant chemotherapy may benefit patients with LMS-B and LMS-ST with tumor size >10 cm, although the statistical significance was not identified because of the limited number of cases, especially in LMS-B cases.

Among bone sarcomas, osteosarcoma and undifferentiated pleomorphic sarcoma of the bone are evidently sensitive to chemotherapy (21). Compared with these bone sarcomas, LMS-B was less chemosensitive; however, our results indicated that adjuvant chemotherapy may benefit patients with localized high-grade LMS-B. For osteosarcoma cases, chemotherapeutic regimens, including methotrexate, cisplatin and doxorubicin with or without ifosfamide, are recommended, although it is unclear which drug should be administered for LMS-B. Among our cases, there were no differences between the outcomes of patients treated with cisplatin-based chemotherapy and doxorubicin- and ifosfamide-based chemotherapy (data not shown), which was consistent with the findings of a previous study (11). Moreover, only a few studies have analyzed the efficacy of adjuvant chemotherapy specifically for LMS-ST. EORTC 62931, a randomized control study of adjuvant chemotherapy for STS, revealed that adjuvant chemotherapy did not improve OS (22). However, a sub analysis of this study indicated that patients with larger and high-grade tumors could benefit from adjuvant chemotherapy (22), which was consistent with our results for LMS-ST. Although the efficacy of adjuvant chemotherapy for LMS-ST is limited, it could be considered for patients with localized high-grade LMS-ST with a tumor size of >10 cm. Our analysis could not identify the suitable regimen and dose of the chemotherapeutic agent, and further studies are needed to determine whether adjuvant chemotherapy is beneficial for patients with LMS-B and LMS-ST. Recent developments have indicated that trabectedin, eribulin and pazopanib can be considered for STS treatment (23). Furthermore, a combination therapy with doxorubicin and trabectedin has been reported to be effective for advanced LMS-ST (24), and adjuvant therapy with these drugs could improve the prognosis of LMS-ST. In addition, in the aforementioned genomic study on LMSs of various locations, some extremity LMSs were classified as ‘dedifferentiated’ subtype, which possesses a high mutation burden and homologous recombination deficiency, and patients may benefit from therapies involving Poly(ADP-ribose) polymerase (PARP) inhibitors and other DNA damage response inhibitors (6).

There are several limitations to this study that should be considered when interpreting the results. First, the number of LMS-B cases was very small, despite the use of data from the Japanese nationwide registry. Second, the database does not include some important factors that may also be associated with patient survival and could influence the results of this study, including detailed chemotherapy dosage, relevant responses and precise anatomical location of the primary tumor. Third, data quality and missing data should be considered when interpreting the results. Fourth, the database does not contain information about prior radiation therapy, which is known to be a risk factor for LMS-B (2). There are few studies about LMS-B, which included at most 74 cases (11–13,16), and these studies analyzed the prognostic factors of LMS-B with metastasis at first presentation and various primary tumors and histological grades. In contrast, we selected cases with high-grade and localized LMS-B and LMS-ST tumors that were resected with curative intent, to evaluate the efficacy of adjuvant chemotherapy. To the best of our knowledge, there are no previous reports focusing on such cases, and this is the first report that indicates the possible benefit of adjuvant chemotherapy for patients with LMS-B and LMS-ST, although further analysis with a larger number of cases and precise information on chemotherapy should be considered.

To conclude, the clinical characteristics and outcomes, including tumor size, grade and location of the primary tumor, were similar for patients with LMS-B and LMS-ST, although patients with LMS-B were of a younger age at first presentation and tended to have a dismal prognosis than patients with LMS-ST. Regarding extrapulmonary metastatic sites, patients with LMS-B tended to have bone metastasis, whereas those with LMS-ST had lymph node metastases. Adjuvant chemotherapy may improve the survival of patients with both LMS-B and LMS-ST with localized high-grade tumors of large size. This analysis could be helpful for clinicians in treating LMS of the bone and soft tissues.

Authors’ contributions

HK and LZ wrote the paper and performed the literature review. YT, MI, TH and ST contributed to the conception and design of the manuscript and critically revised the manuscript. All authors read and approved the final version of the manuscript.

Conflict of interest statement

None declared.

Funding

Not applicable.

References