Amputation in patients with extremity soft tissue sarcoma: the experience of an East Asian referral center

In 1960s, amputation was recommended as the treatment for STS with high local recurrence rate [3]. LSS with multidisciplinary treatment has begun to replace amputation since 1970s [2, 14, 21]. However, amputation still sustains its role as a definitive treatment to achieve local control, and many studies have revisited its contemporary indication and outcome recently [13, 22‐24]. Over the years, STS patients showed reluctance to undergo radical surgery such as amputation. The amputation rate for localized STS reported in our study was similar but slightly lower than that reported in Western studies, and the rate of amputation with primary curative intent was lower (1.8%). The DSS and MFS rates of amputation for localized disease reported in this study were higher than that indicated in previous studies, while the LRFS rates were similar.

Our study has some limitations. First, this study evaluated a small number of extremity STS patients from a single tertiary referral center, which might be subject to inherent selection bias and referral bias. In the subgroup analysis of the PL and RL groups, the analysis was inevitably descriptive without statistical confirmation due to the small sample size. Future multi-center study or national level study is necessary to re-evaluate or confirm our results. Secondly, due to its retrospective study design, it was difficult to assess the reason of low amputation rate in our institute. We admit that there might have been patients who needed amputation were treated with LSS and adjuvant treatments. We instead tried to overcome the limitation by analyzing the indications for amputation in the PL and RL group. We also compared the amputation rate of our cohort with those reported in the literature. On the other hand, we compared the oncologic outcomes between the amputation group and LSS group to assess whether this low rate of amputation might have failed in achieving proper local control. There might have been multiple factors influenced avoidance of amputations including preference of the surgeon, cultural issues such as the Confucianism. We believe these factors should be assessed by analyzing the patients who did not undergo amputation and underwent LSS with adjuvant treatment in the future.

Thirdly, our study could not analyze socioeconomic status (SES), which is important for health care accessibility. Some researchers suggest that SES is more predictive of treatment quality received than race or ethnicity [20, 25]. However, this element varies depending on countries and Korea has a universal health insurance system with short waiting times while accessibility to tertiary referral centers is high [26]. In addition, this study is a review of single institute. Therefore, consideration for this status is also necessary in future multi-center design or National level study.

The amputation rate for localized disease was 3.6% (22/596), while that with primary curative intent was 1.8% (9/490). Both of the values were similar or slightly lower than those indicated in previous Western studies (Table 2). In Europe, according to the most recent study of Royal Marsden Hospital with 556 STS patients, Smith et al. reported that the amputation rate with primary curative intent was 4.1% [13]. Keus et al. reported 8.3% of amputation rate from Netherland [27]. Trovik et al. reviewed 1,613 adult patients who were registered to the Scandinavian Sarcoma Group Registry and reported a localized amputation rate of 22.0% with a primary amputation rate of 9.6% [12]. In Canada, Ghert et al. reported a primary amputation rate of 6.3% in 413 patients [15]. In the United states, Williard et al. at the Memorial Sloan-Kettering Cancer Center reported 14.1% of amputation rate with 557 STS patients.[28]. Twenty years later, Alamanda et al. reviewed 278 patients with STS and reported a localized amputation rate of 5.7% and a primary amputation rate of 3.5% [29]. The amputation rate has significantly decreased over the years and the rate in our institute was similar or slightly lower than other references. These numbers also show that there are certain indications that the amputation is necessary.

In clinico-pathologic factors, the median tumor size was less than 5 cm and smaller than those of previous studies. Erstad et al. reported that 69% of amputated tumors were larger than 5 cm, while 34% were larger than 10 cm. Stevenson et al. reported median tumor sizes of 8.7 cm in the entire cohort and 11.5 cm in the primary localized group [22, 23]. The pathologic grades and depth in our cohort were similar with those in previous studies. Thus, we carefully assumed that less aggressive tumor biology with small size tumor might have contributed to the favorable oncologic outcome of our cohort. No difference was observed in the oncologic outcome between the PL group and RL group. Smith et al. reported the same phenomenon in his retrospective review of 69 patients who underwent amputation [13]. Therefore, Smaller tumors might have been the explanation for good prognosis of the patients with amputation. We believe there might have been other factors affected the patients with aggressive large sarcomas to betreated with LSS instead of amputation. This should be analyzed in the future study. When the PL and RL groups were compared, there were no difference in tumor size between the two groups. In Stevenson et al.’s recent study, the median tumor size of the primary amputation group (11.5 cm) was larger than that of the group who underwent amputation as non-primary treatment (6.1 cm) (p = 0.008). Therefore, the size was not the definitive factor for deciding amputation in our study. Erstad et al. classified the tumors located in the hand and foot as the distinctive group and suggested that this group had a better overall survival [22]. In our PL group, two patients had foot tumors and one had a hand tumor. All of them were successfully treated with amputation. Hand and foot tumors were not observed in the RL group. On histologic grade, grade 3 was the most common in the PL group (44.4%) and grade 2 was the most common in the RL group (61.5%). This result is similar to that reported in Smith et al.’s review of 69 STS patients treated with major amputation [13], but slightly different from that reported in Erstad et al’s review, in which grade 3 tumors were the most common histology in both groups. Even though grade 2 was more frequent in RL group, RL tumors are recurrent tumors which have survived the selective pressure of LSS and this point should be considered in treatment strategy [22]. Multifocality was a characteristic of RL tumors and statistically different from that of PL tumors (p = 0.031). Approximately 62% of RL tumors had multifocal location. Stojadinovic et al. suggested multiple local recurrence and multi-compartment disease as principal reasons for amputation [30]. Even a single lesion can involve multiple compartments. Multifocal recurrent lesions usually involve multiple compartments (100% in the RL group) and entail inevitable functional loss in an attempt to achieve R0 margin. Therefore, multifocality of recurrent STS can be considered as a distinctive factor for considering amputation.

\In relation to the oncologic outcomes, the DSS and MFS rates in our study were higher than those in previous studies, while the LRFS rate was compatible with that in Western studies (Table 4). Most previous studies reported DSS or OS rates of 40–50% with similar MFS rates. Although the mode of surgery is not a predictive factor for metastasis, tumors requiring amputation tend to have aggressive biology such as larger size, higher grade, and deep location in literatures. [31‐33]. Japanese report of Tokai musculoskeletal consortium with 55 patients showed 52.8% of DSS and 63.1% with 42 patients without metastasis at presentation [18]. In this study, they have excluded amputations distal to wrist and ankle which might have influence the results while there were 3 amputations distal to wrist and ankle in our study.