Efficacy of surgery in the management of multiple recurrences of retroperitoneal dedifferentiated liposarcoma

Abstract

Background

Retroperitoneal dedifferentiated liposarcoma is associated with a high risk of recurrence; however, treatment strategies that are more effective than surgery remain to be established. This study aimed to determine the optimal number of surgeries that would be effective for patients with recurrent disease. Furthermore, the improvement in prognosis was evaluated according to the malignancy level.

Methods

The effect of each type of surgery on the prognosis of 118 patients with retroperitoneal dedifferentiated liposarcoma treated at the Osaka International Cancer Institute between 1997 and 2022 was investigated. Among the 118 patients, 103 underwent initial surgery, while 54 and 30 patients underwent second and third surgeries, respectively. The overall and disease-free survival rates of each group were compared using the Kaplan–Meier method, and the log-rank test was used to determine statistical significance in univariate analysis. ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) was used to assess malignancy. Maximum standardized uptake values (SUVmax) of ≥ 4 and < 4 were classified as high and low malignancy, respectively.

Results

The first and second surgeries resulted in a significant improvement in the overall survival rate, regardless of the malignancy level (p < 0.001); however, no significant improvement in prognosis was observed after the third surgery (p = 0.077). Low-grade malignancies are associated with a better postoperative prognosis, even in cases of recurrence. In contrast, high-grade malignancies exhibit a reduction in surgical efficacy.

Conclusions

This study highlights the importance of considering the tumor malignancy level and the patient's overall condition when deciding whether to perform repeated surgical interventions. Surgical treatment can prolong overall survival, even in patients with recurrence; however, it is advisable to assess malignancy levels when determining the suitability of surgery beyond the second recurrence.

Introduction

Dedifferentiated liposarcoma (DDLPS), a rare and aggressive liposarcoma subtype that originates from fat cells, is a type of soft tissue sarcoma [1]. DDLPS is characterized by the presence of well-differentiated liposarcomas and non-lipogenic sarcomatous components. The dedifferentiated component of DDLPS is of a higher grade, less differentiated, and more aggressive than the well-differentiated component. The retroperitoneum (area behind the abdominal cavity) is the most common site for DDLPS. DDLPS may not present with any accompanying clinical symptoms; consequently, tumors > 10 cm in size are commonly detected. The amplification of MDM2 and CDK4 (chromosome 12 q13-15 region) has been implicated in DDLPS incidence; however, the exact cause of dedifferentiation remains unclear [2,3,4,5,6].

Currently, the curative treatment for retroperitoneal sarcomas is limited to surgery. However, the risk of recurrence is high even in cases with wide resection margins, posing a clinical challenge. Recurrence is often associated with increased surgical complexity owing to postoperative adhesions and other factors, and the median follow-up duration from recurrence of retroperitoneal sarcoma was 41 months in a previous study [7]. The median overall survival (OS) after local recurrence only, distant metastasis only, and local recurrence and distant metastasis were 33, 25, and 12 months, respectively [7]. Patients with DDLPS require careful management and monitoring by a team of medical professionals, including oncologists and surgeons because of their aggressive nature and metastatic potential. Moreover, the decision to perform surgery in patients with recurrence plays a crucial role in determining prognosis. However, few studies have investigated the efficacy of surgery in the management of recurrent liposarcoma; consequently, its effectiveness remains unclear. This study investigated the effect of the number of surgeries on the prognosis of patients with recurrent DDLPS, and assessed postoperative complication rates for each surgical procedure. Liposarcomas with higher maximum standardized uptake values (SUVmax) calculated using ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) have been considered to be more malignant and associated with poorer prognosis in previous studies [8]. Therefore, this study aimed to elucidate whether reoperation decisions based on SUVmax values for malignancy classification were effective.

Methods

A total of 118 patients (43 men and 75 women) with retroperitoneal DDLPS who presented to the Osaka International Cancer Institute between 1997 and 2022 participated in this study. The median age and follow-up period were 63.8 years and 53.6 months (range, 0.8–307.8 months), respectively. The 84, 31, and three patients had grade II, III, and unclassified DDLPS, respectively. Five patients, four of whom underwent surgery, presented with distant metastases, and the median tumor size was 13.6 cm (range: 2–45 cm) (Table 1). The completeness of the primary tumor resection was classified dichotomously as macroscopically complete (R0/1) or incomplete (R2) according to operative and pathological reports. The relationship between malignancy and number of surgical procedures performed was also evaluated. Patients were divided into two groups based on the SUVmax values: Group 1 (aggressive group; SUVmax values: ≥ 4) and Group 2 (non-aggressive group; SUVmax values: < 4).

Table 1 Clinical characteristics of the 118 patients with DDLPS

The OS and disease-free survival (DFS) rates of each group were determined using the Kaplan–Meier method, and the log-rank test was used to determine statistical significance in the univariate analysis. OS-0 was defined as the interval between initial presentation and mortality due to the disease or the last follow-up visit. OS-1/OS-2/OS-3 was defined as the interval between the detection of recurrence before the second, third, or fourth surgery and mortality caused by the disease or the last follow-up visit. DFS-0 was defined as the interval between the first surgical resection and the first detection of local recurrence or distant metastasis, or the last follow-up. DFS-1 and DFS-2 were defined similarly, and patients who presented with distant metastasis at the time of initial referral and those who did not undergo definitive surgery were excluded from the progression-free survival (PFS) analysis. Statistical significance was set at p < 0.05.

Tumor grade was evaluated using the FNCLCC system (Fédération Nationale des Centres de Lutte Contre le Cancer system), which is a pathological grading system used to evaluate the grade of soft tissue tumors such as sarcomas [9]. It assesses tumor malignancy based on three key factors: differentiation, mitotic count, and tumor necrosis. The scores from these factors were combined to determine the overall grade of the tumor, ranging from Grade I (low grade) to Grade III (high grade). This system helps predict tumor behavior and guides treatment decisions.

We evaluated the incidence and severity of postoperative complications using the Clavien–Dindo classification system, which is a grading system for postoperative complications based on the required level of intervention [10]. Grade I complications included minor complications requiring minimal treatment. Grade II included complications requiring pharmacological treatment. Grade III encompasses complications requiring surgical, endoscopic, or radiological intervention, with IIIa not requiring general anesthesia and IIIb requiring general anesthesia. Grade IV includes life-threatening complications requiring intensive care, with IVa for single-organ dysfunction and IVb for multiple-organ dysfunction. Finally, Grade V denotes complications leading to death. This system standardizes complication reporting and facilitates comparison of outcomes across different treatments and institutions [11].

Furthermore, the Evans–Pridon classification is frequently used for the pathological diagnosis of sarcomas [12]. The Evans–Pridon classification system is used to evaluate the pathological response of tumors to chemotherapy, particularly in the context of osteosarcoma. This classification is based on histological changes observed in the tumor tissue following treatment and is commonly employed to assess the effectiveness of chemotherapy. In this study, FNCLCC grades were extracted from patient records based on diagnoses made by multiple physicians. Additionally, because chemotherapy and radiation therapy were administered only in patients with advanced unresectable retroperitoneal liposarcoma at our institution, the Evans-Pridon classification could not be evaluated.

Intergroup differences were compared using the Mann–Whitney U or Fisher’s exact tests. Statistical significance was set at p < 0.05. All statistical analyses were performed using EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). This study was approved by the Institutional Review Board of Osaka International Cancer Institute.

Results

Among the 118 patients, 103 underwent the initial surgery. Surgery was deemed impossible in 15 patients owing to size (n = 7), presence of aortic infiltration (n = 5), distant metastasis (n = 1), refusal to undergo surgery (n = 1), and other complications (n = 1). Among 77 patients (75%) who experienced recurrence after the initial surgery, 54 underwent a second surgery, and 51 patients (94%) experienced recurrence after the second surgery, 30 of whom underwent a third surgery. Twenty-three (77%) patients experienced recurrence after the third surgery, 17 of whom underwent a fourth surgery (Fig. 1). The decision to perform surgery was made by multiple experienced orthopedic surgeons who specialize in musculoskeletal tumors as well as skilled general surgeons and urologists. The medical professionals decided to proceed with the surgery only after considering the patient’s overall health, location of the tumor, and its local extent and determining that it was safe to do so. Most patients had a performance status of 0 or 1, and there was no difference between the surgical and non-surgical groups. Figure 2 in the provided document presents a series of Kaplan–Meier survival curves that illustrate the OS probabilities of patients over time, measured in months. The figure is divided into four parts, labeled A, B, C, and D, each corresponding to a different stage or subgroup. A significant improvement in OS was observed in the group of patients who underwent initial surgery, compared to the non-surgical group (p < 0.001) (Fig. 2A). The median OS-0 of the surgery and non-surgery groups was 147 and 32 months, respectively. The 54 patients who underwent a second surgery were included in the surgery group, and the 23 patients who did not undergo a second surgery were included in the non-surgery group. The median OS-1 scores in the surgical and non-surgical groups were 111 and 17 months, respectively, and a statistically significant improvement in OS was observed in the group that underwent a second surgery (p < 0.001) (Fig. 2B). Thirty patients who underwent a third surgery were included in the surgery group, and 21 patients who did not undergo a third surgery were included in the non-surgery group. The median OS-2 scores in the surgical and non-surgical groups were 68 and 32 months, respectively. The prognoses of these groups tended to improve after surgery; however, the difference was not statistically significant (p = 0.077) (Fig. 2C). Seventeen patients who underwent a fourth surgery were included in the surgery group, and the six patients who did not were included in the non-surgery group. The median OS-3 score of the surgery group was 47 months, and while the median OS-3 of the non-surgery group was not determined, no statistically significant differences were observed between the groups (p = 0.916) (Fig. 2D). These findings indicate that a second surgery should be performed to improve the prognosis of patients with recurrent liposarcoma. Organs resected concurrently with the wide tumor resection were also evaluated (Table 2). Concurrent resection of the renal and urinary organs, observed in 45 patients, was performed most frequently during the initial surgery. For the 50 patients with a second recurrence, we compared OS and DFS between the surgery and surgery plus radiation therapy groups. No significant differences were observed in the OS (p = 0.567) or DFS (p = 0.336). Similarly, we compared the surgery group with the surgery plus chemotherapy group; however, no significant differences were observed in terms of OS (p = 0.834) or DFS (p = 0.48). Of the 25 patients who underwent a third surgery, eight received radiation therapy and 15 received chemotherapy. We investigated the OS in both groups, but no significant differences were observed (p = 0.564, p = 0.873).

Fig. 1

Surgical flowchart depicting the inclusion of 118 patients with DDLPS. DDLPS, dedifferentiated liposarcoma

Fig. 2

OS of patients compared to the presence or absence of surgery. A OS-0, (B) OS-1, (C) OS-2, and (D) OS-3. OS, overall survival

Table 2 Combined resected organs per surgery

The OS and DFS rates between the two groups (Group 1: aggressive and Group 2: non-aggressive) were compared based on the number of surgeries performed. The FNCLCC Grade of Group 1 was significantly worse, and the number of patients who underwent a second surgery was significantly higher. No significant differences were observed in the resection margins per surgery (Table 3), and Fig. 3 in the provided document compares survival probabilities between the two groups across different time points. The figure is divided into parts labeled A and B, each showing the survival probability (OS) over time in months. The OS-0 and OS-1 scores of patients in Group 1 who underwent the first and second surgeries were significantly better; however, no significant differences were observed in the OS beyond the second operation (OS-2) (Fig. 3A). The OS-0, OS-1, and OS-2 scores of patients in group 2 who underwent surgery were significantly better (Fig. 3B). Figure 4 in the document appeared to display Kaplan–Meier curves illustrating disease-free survival (DFS) over time, measured in months. The figure was divided into three parts labeled A, B, and C, each showing the DFS probabilities under different conditions or for different groups. Figure 4A (DFS-0) shows the DFS probability with a p value of 0.52. Figure 4B (DFS-1) shows a statistically significant difference in DFS, with a p-value of less than 0.05. Figure 4C (DFS-2) also shows a significant difference in DFS, with a p-value of less than 0.01, indicating a strong statistical difference in disease-free survival between the groups at this stage. DFS-1 and DFS-2 levels in Group 1 were significantly lower than those in Group 2 (p < 0.05); however, no significant differences were observed between the two groups in DFS-0 (Fig. 4A–C). The Clavien-Dindo classification was used to assess early postoperative complications for each surgery. According to our hospital’s internal regulations, all cases of retroperitoneal tumor surgery require admission to the ICU postoperatively. None of the patients required intensive care. Because postoperative ICU admission may correspond to grade 4, patients who were discharged from the ICU the day after surgery were excluded. In the present study, all early postoperative complications were Grade I or II. No Grade III or higher complications, including mortality, were observed within 30 days after surgery (Table 4).

Table 3 Clinical characteristics of patients with aggressive or non-aggressive DDLPS

Fig. 3

OS of the patients in Group 1 (aggressive; SUVmax ≥ 4) and Group 2 (non-aggressive; SUVmax < 4). A Group 1: OS-0, OS-1, OS-2. B Group 2: OS-0, OS-1, OS-2. OS, overall survival; SUVmax, the maximum standardized uptake value

Fig. 4

DFS of the patients in Group 1 (aggressive; SUVmax ≥ 4) and Group 2 (non-aggressive; SUVmax < 4). A DFS-0, (B) DFS-1, and (C) DFS-2. DFS, disease-free survival; SUVmax, the maximum standardized uptake value

Table 4 Clavien-Dindo classification per surgery

Discussion

Resection of retroperitoneal DDLPS with adequate margins is often challenging because of its anatomical characteristics and the presence of important surrounding organs, which result in frequent recurrence. It is crucial that both the tumor and any potentially infiltrated adjacent structures are removed en bloc in a single session [13,14,15]. Bonvalot et al. reported that patients diagnosed with retroperitoneal sarcoma who underwent surgery between 2000 and 2009 over a nine-year period had a high long-term survival rate and a low rate of local recurrence [16]. In particular, favorable outcomes were observed for small or low-grade tumors.

The findings of the present study suggest that a third surgery significantly increases OS and PFS in patients with non-aggressive (SUVmax values: < 4) DDLPS. Patients who undergo complete (macroscopic) or compartmental resection (R0 or R1) of the primary tumor exhibit an improved prognosis, with a 5-year OS rate of 54–70%; however, local recurrence is observed in 41–50% of these cases within 5 years of surgery [17, 18]. The prognosis of patients with retroperitoneal sarcoma with SUVmax values of ≥ 4 was poorer than that of the patients with SUVmax values of < 4. A comparison of the 84 patients who underwent FDG-PET and were evaluated using SUVmax values revealed that the prognosis was significantly favorable in the low-malignancy group (Supplemental Fig. 1). This finding is supported by the fact that recurrence occurred within two years in most patients (92%) who experienced a second recurrence. Wakamatsu et al. examined the characteristics of low and high SUVmax groups in DDLPS [8]. In the low SUVmax group, the majority of tumors were classified as FNCLCC grade II, whereas 45.5% (15 of 33) of tumors in the high SUVmax group were classified as FNCLCC grade III. Because tumor grade is a significant prognostic factor, they hypothesized that the shorter prognosis observed in patients with a high SUVmax could be attributed to the higher incidence of grade III tumors. However, the prognosis of patients with a high SUVmax and grade II tumors is poorer than that of patients with a low SUVmax and grade II tumors. These findings suggest that SUVmax has potential clinical utility in refining the prognostic assessment of patients with DDLPS, particularly those with grade II tumors.

Local recurrence remains common even after optimal resection of retroperitoneal liposarcomas, and is the most frequent cause of death. Adjuvant radiation therapy (RT) and chemotherapy may be valuable treatment options for improving local control, particularly in patients with margins of involvement or high-grade tumors. The EORTC Surgery With or Without Radiotherapy for Retroperitoneal Sarcoma (STRASS) trial evaluated the role of preoperative radiotherapy in the treatment of retroperitoneal sarcomas, including DDLPS [19]. In total, 266 patients with retroperitoneal sarcoma, including those with dedifferentiated liposarcoma, were enrolled. These patients were randomized into two groups: one receiving preoperative radiotherapy followed by surgery and the other undergoing surgery alone. The results of the STRASS trial indicated that neoadjuvant radiotherapy did not significantly improve local control or overall survival in patients compared with surgery alone. Specifically, the benefits of radiotherapy appear to be limited for dedifferentiated liposarcomas. However, the trial findings suggested that individual patient and tumor characteristics might influence the effectiveness of radiotherapy, highlighting the need for personalized treatment approaches. In addition, several small trials have shown marked variations in the RT dose, fractionation, concurrent use of chemotherapy, delivery method (external beam or brachytherapy), timing (preoperative, intraoperative, or postoperative), and energy carriers (photons, electrons, protons, or carbon ions) [14].

Chemotherapeutic drugs, such as anthracyclines (doxorubicin) and the alkylating agent ifosfamide, have been used in many cases. Eribulin, trabectedin, and pazopanib were established as effective second- or third-line options for treating resistant disease [20,21,22]. Several reports have suggested an efficacy rate of 10–20% [23,24,25,26]. The STRASS-2 trial is an ongoing Phase III clinical study aimed at determining whether preoperative (neoadjuvant) chemotherapy can improve outcomes in patients with high-risk retroperitoneal sarcoma, specifically focusing on subtypes such as DDLPS and leiomyosarcoma [27]. The trial is being conducted in multiple countries, including Europe, the USA, Canada, Japan, and Australia, with a target enrollment of 200 patients. The goal of this study was to assess whether adding chemotherapy before surgery enhances disease-free survival compared to surgery alone. Recently, clinical trials on MDM2 inhibitors have indicated that they may be effective treatment option [28]. Evaluation of the effects of RT and chemotherapy in 50 patients with second recurrence revealed no significant differences. However, performing RT and chemotherapy in addition to surgery tended to improve the prognosis, indicating that it may aid in decision-making based on the situation.

The rate of major postoperative complications in major series was 15–20% [29, 30]. Among 1007 patients with primary retroperitoneal sarcoma, a serious postoperative complication (Clavien-Dindo ≥ 3) was observed in 16.4% of patients, 10.5% required reoperation, and 1.8% died within 30 d of surgery in a previous study [29]. In the present study, the rate of Grade II complications increased with each additional surgery; however, no Grade III or higher complications were observed. It can be inferred that the favorable outcomes were likely attributable to meticulous postoperative systemic management and effective coordination with physicians and nurses from other departments. Intravenous patient-controlled analgesia or hydration therapy during fasting has been used to manage postoperative pain in almost all patients. Factors contributing to the increase in Grade II complications include securing central venous access and blood transfusions.

The present study had some limitations that should be acknowledged. First, because it was conducted at a single institution, the generalizability of the findings may be limited. Differences in patient demographics, treatment protocols, and diagnostic practices at other centers may affect the applicability of our results to a broader population. Second, being a retrospective study, it is inherently subject to selection and information biases. Data collection relied on existing medical records, which may not have captured all relevant variables or may have included inaccuracies. Third, potential confounding variables such as variations in treatment adherence, patient comorbidities, and other external factors were not accounted for in the analysis. These factors could influence the study outcomes and introduce additional sources of bias.

Conclusions

A second excision leads to significantly improved prognosis in patients with recurrent retroperitoneal liposarcoma. However, surgeries after the third recurrence offer limited prognostic improvement in patients with highly malignant lesions, necessitating careful consideration of their appropriateness.