2448
Retroperitoneal Soft Tissue Sarcomas Patterns of Recurrence in 167 Patients Treated at a Single Institution
Alessandro Gronchi, M.D.1 Paolo G. Casali, M.D.1 Marco Fiore, M.D.1 Luigi Mariani, M.D.2 Salvatore Lo Vullo, B.Sc.2 Rossella Bertulli, M.D.3 Maurizio Colecchia, M.D.4 Laura Lozza, M.D.3 Patrizia Olmi, M.D.3 Mario Santinami, M.D.1 Juan Rosai, M.D.4 1
Department of Surgery, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy.
2
Department of Biostatistics, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy.
3
Department of Radiation Oncology, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy.
4
Department of Pathology, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milano, Italy.
BACKGROUND. The objective of this study was to assess long-term prognosis and patterns of failure in patients with retroperitoneal soft tissue sarcoma who underwent surgery with curative intent at a single institution. Several series of patients with retroperitoneal sarcoma were reported, providing widely variable data regarding local and distant failure. METHODS. Overall, 167 consecutive patients who underwent surgery for retroperitoneal soft tissue sarcoma over a 20 year span at a single referral center were reviewed retrospectively. Eighty-two patients presented with primary disease, whereas 85 patients had recurrent locoregional tumors. Surgical resection was considered macroscopically complete in 147 of 167 patients (88%). RESULTS. Overall survival at 10 years after definitive surgery was approximately 27%, and the disease-free survival was approximately 16%. Only a minority of patients developed distant metastases. The risk of recurrence was correlated with whether patients underwent surgery for primary disease or for recurrent disease; the 10 year disease-free survival rate was 27% in the former group and 4.6% in the latter group. Histotype and malignancy grade were other prognostic factors, with the former found to be predictive of the pattern of failure and the latter predictive of overall survival. CONCLUSIONS. Local recurrence after primary surgery and high-grade malignancy were associated with the worst survival. Histologic subtype appeared to influence the pattern of recurrence, which mainly was local for patients with liposarcoma but was both local and distant for patients with the other histotypes. Cancer 2004;100: 2448 –55. © 2004 American Cancer Society.
KEYWORDS: sarcoma, retroperitoneal tumors, prognosis, histotype, pattern of failure.
S
Address for reprints: Alessandro Gronchi, M.D., Department of Surgery, Istituto Nazionale per lo Studio e la Cura dei Tumori, via Venezian,1-20133, Milano, Italy; Fax: (011) 39 0223902404; E-mail:
[email protected] Received November 7, 2003; revision received February 2, 2004; accepted March 12, 2004.
oft tissue sarcomas (STS) are rare tumors that comprise only 1% of all malignancies.1 The retroperitoneum is an uncommon site of origin for them, accounting for 10% of all STS. Because of its anatomic peculiarities (lack of boundaries to allow true compartmental surgical resection and asymptomatic growth of huge masses), the presentation of STS remains challenging. The prognosis of patients with STS generally is poor, especially when they are assessed over the long term.2–10 The contribution of adjuvant chemotherapy to the treatment of adult patients with STS remains questionable, whereas radiation therapy in patients with retroperitoneal STS is either unfeasible or is affected strongly by anatomic constraints (mainly related to the presence of kidneys and small bowel). Therefore, the main therapeutic burden is placed on surgery.2–10 However, surgery alone is curative in only a relatively small number of patients. Available studies vary widely with regard to estimates of cured or long-term survivors, most likely depending on strong variations in the referral pattern of the
© 2004 American Cancer Society DOI 10.1002/cncr.20269 Published online 26 April 2004 in Wiley InterScience (www.interscience.wiley.com).
Patterns of Recurrence in Retroperitoneal STS/Gronchi et al.
various institutions. Indeed, the approach to retroperitoneal STS is under active investigation. Novel radiation therapy techniques have been explored,11–21 and the role of adjuvant chemotherapy still is debated.22,23 However, efforts to improve treatment results should be based on the patterns in which these tumors currently fail. Thus, the objective of this retrospective analysis was to assess long-term prognosis and patterns of failure in a large cohort of patients with retroperitoneal STS who underwent surgery with curative intent at a single institution over 20 years.
MATERIALS AND METHODS Between January 1982 and December 2001, 256 patients with retroperitoneal primary tumors were referred for treatment to the Istituto Nazionale per lo Studio e la Cura dei Tumori, Milan, Italy. Among these, 20 patients were affected by retroperitoneal aggressive fibromatosis (desmoid tumors), 15 patients had benign disease, 11 patients had adrenal neoplasms, and 5 patients were diagnosed with desmoplastic small round cell tumors. All these patients were excluded from the current analysis. Two hundred five patients were proven to have retroperitoneal, adulttype STS. Thirty-one patients had lung metastases at the time of presentation and, thus, also were excluded from the current series. Thus, 174 patients were treated with curative intent. Seven patients were lost to follow-up soon after surgery, and they also were excluded from this analysis. The outcome of the remaining 167 patients is the subject of this article. Patient characteristics are summarized in Table 1. The median patient age was 53 years (range, 15– 82 years). Seventy-four patients were females and 93 patients were males, for a 1.2:1 male-to-female ratio. Liposarcoma was the main histotype (96 patients), followed by leiomyosarcoma (28 patients), malignant fibrous hystiocytoma (14 patients), malignant peripheral nerve sheath tumor (13 patients), hemangiopericytoma (5 patients), and other sarcomas (11 patients). Fifty-nine patients (35%) were classified with Grade 1 tumors, 44 patients (26%) were classified with Grade 2 tumors, and 64 patients (39%) were classified with Grade 3 tumors according to Federation Nationale des Centres de Lutte Contre la Cancer (FNCLCC) criteria.24 Eighty-two patients presented with primary disease and 85 patients presented with recurrent locoregional tumors. Surgical resection was adjudged macroscopically complete in 147 of 167 patients (88%). Radiation therapy was delivered as an adjunct in 49 patients, including 27 patients with primary disease and 22 patients with recurrent disease. The decision to administer radiation therapy was taken jointly by the operating
2449
TABLE 1 Main Patient and Disease Characteristics According to Tumor Presentation No. of patients (%) Characteristic
Primary tumor
Recurrent tumor
Overall
Total no. Age (yrs) Median Range Gender Female Male Tumor size ⱕ 10 cm 10–20 cm. ⬎ 20 cm NA Histology Liposarcoma Leiomyosarcoma MFH MPNST Hemangiopericytoma Other Tumor grade Low (Grade 1) High (Grade 2 or 3) RT Not done Done CT Not done Done
82 (100.0)
85 (100.0)
167 (100.0)
53 15–77
53 33–82
53 15–82
41 (50.0) 41 (50.0)
33 (38.8) 52 (61.2)
74 (44.3) 93 (55.7)
27 (33.3) 23 (28.4) 31 (38.3) 1
29 (34.5) 32 (38.1) 23 (27.4) 1
56 (33.9) 55 (33.3) 54 (32.7) 2
33 (40.2) 20 (24.4) 7 (8.6) 11 (13.4) 5 (6.1) 6 (7.3)
63 (74.2) 8 (9.4) 7 (8.2) 2 (2.4) — 5 (5.9)
96 (57.5) 28 (16.7) 14 (8.4) 13 (7.8) 5 (3.0) 11 (6.6)
28 (34.1) 54 (65.9)
31 (36.5) 54 (63.5)
59 (35.3) 108 (64.7)
55 (67.1) 27 (32.9)
63 (74.1) 22 (25.9)
118 (70.6) 49 (29.3)
68 (82.9) 14 (17.1)
65 (76.5) 20 (23.5)
133 (79.6) 34 (20.4)
NA: not available; MFH: malignant fibrous histiocytoma; MPNST: malignant peripheral nerve sheath tumor; RT: radiotherapy; CT: chemotherapy.
surgeon and the radiation oncologist when they believed there was a high risk of local recurrence based on clinical grounds. However, no prospectively stated, objective criteria were used to this end. External beam radiation (in a preoperative setting for 19 patients and in a postoperative setting for 30 patients) was given at a dose ranging from 36 – 65 grays (Gy) (median, 50 Gy). Chemotherapy was given to 34 patients, including 13 patients who had low-grade/intermediate-grade STS (Grade 1 and 2) and 21 patients who had had high-grade STS. Anthracycline-based regimens were used to treat all of these patients, according to standard protocols used at the time. The decision to administer chemotherapy was taken jointly by the medical oncologist and the operating surgeon with the hope of reducing the chances of local failure. The median follow-up for the entire group, as of September 2002, was 65 months (interquartile range, 21–154 months).
2450
CANCER June 1, 2004 / Volume 100 / Number 11
Statistical Methods Overall survival was calculated from the time of diagnosis to the time of death or of the last recorded follow-up for living patients. To analyze survival, a delayed-entry approach was adopted for patients who underwent surgery for recurrent locoregional disease: observation times for these patients were “left truncated” at the time of surgery,(i.e., they were not considered among the patients who were at risk during the interval between diagnosis and curative surgery at the study institute); however, their time zero was left at the time of diagnosis. Such a procedure was adopted to avoid the time-length selection bias that would arise otherwise, because patients who tend to live longer have a greater chance of entering the study cohort. Analysis of overall survival was performed using a Cox multiple regression model. Putative prognostic factors that were included in the model were tumor size (⬍ 10 cm, 10 –20, or ⬎ 20 cm in maximum dimension; 3 classes were allowed to investigate a possible trend for tumor size), tumor grade (Grades 1–3), and histotype (liposarcoma or other). Disease presentation (primary or recurrent) was included as an adjustment factor to account for the possibly diverse meaning of the above tumor characteristics in primary and recurrent tumors. The time to local or distant failure, whichever occurred first, was calculated from the time of surgery at the study institution. Crude cumulative incidence curves for local and distant failures were estimated as described by Marubini and Valsecchi25; in such an analysis, the above-described neoplastic events and deaths in patients who had no evidence of disease were regarded as competing events. Differences between the cumulative incidence curves according to patient subgroups were tested for statistical significance using the procedure of Gray.26 All P values reported are two-sided. P values ⬍ 0.05 were considered significant.
RESULTS Overall, 78 patients underwent simple surgical excision of the mass, and 89 patients underwent extended procedures with resection of visceral organs surrounding the mass in addition to complete excision of the mass. In particular, 47 patients underwent en bloc resection of only 1 organ with the tumor, 26 patients underwent en bloc resection of 2 organs, 10 patients underwent en bloc resection of 3 organs, 4 patients underwent en bloc resection of 4 organs, and 2 patients underwent en bloc resection of 5 organs. The organs resected most frequently were the kidneys followed by the colon, spleen, and pancreas.
FIGURE 1. Overall survival by grade in the entire case series (167 patients). Solid line: low-grade tumor; dashed line: high-grade tumor.
A total of 90 deaths were recorded. Of these, 82 deaths were due to disease progression (56 patients with inoperable locoregional recurrences and 26 patients with distant metastasis), 6 deaths were due to surgical complications, and 2 deaths were due to causes unrelated to the disease. Overall survival for the entire series was 53.6% (95% confidence interval [95%CI], 44.9 – 64.0%) at 5 years and 26.9% (95%CI, 19.7–36.8%) at 10 years. The median survival was 62 months. A trend toward shorter survival for patients with a higher grade of malignancy was observed (Fig. 1). Ninety-nine patients developed recurrences with both local and/or distant disease. Seventy-three patients had only local recurrences after they underwent their first surgical procedure at the study institution. Forty-seven patients had only 1 local recurrence, 16 patients had 2 recurrences, and 10 patients had ⱖ 3 recurrences. Twelve patients had only distant recurrences after undergoing their first surgical procedure at the study institution, involving the lung in 6 patients, the liver in 5 patients, the peritoneum in 2 patients, and the skeletal system in 2 patients. The remaining 14 patients had both local and distant recurrences, which were synchronous with the first local recurrence in 6 patients and metachronous in the remaining 8 patients. In subsequent analyses, synchronous local and distant recurrences were pooled with distant recurrences. The disease-free survival rate for the entire series was 27.6% at 5 years (95%CI, 20.0 –35.3%) and 16.0% at 10 years (95%CI, 8.9 –23.1%), with a better outcome for patients who underwent primary excision versus surgery for recurrence (Fig. 2). The 5-year crude cumulative local recurrence rate was 54.3% (95%CI, 46.0 –
Patterns of Recurrence in Retroperitoneal STS/Gronchi et al.
2451
TABLE 2 Hazard Ratio Estimates with 95% Confidence Intervals and P Values from the Stratified Cox Proportional Hazard Model for Overall Survival Category (reference) Presentation Recurrence (primary) Tumor size 10–20 cm (ⱕ 10 cm) ⬎ 20 cm (ⱕ 10 cm) Tumor grade Grade 2 (Grade 1) Grade 3 (Grade 1) Histology Liposarcoma (other)
FIGURE 2. Disease-free survival by presentation status in the entire case series (167 patients). Solid line: primary tumor; dashed line: recurrent tumor.
HR
95% CI
P valuea
1.45
0.92–2.29
0.1092
1.36 1.11
0.80–2.33 0.62–2.00
0.5153 —
1.89 2.31
1.00–3.57 1.28–4.16
0.0183 —
1.54
0.85–2.80
0.1520
HR: hazard ratio; 95% CI: 95% confidence interval. a Likelihood ratio P value for testing the overall significance of each covariate.
To determine whether the effect of tumor size, grade, and histology differed for primary tumors and recurrent tumors, interactions between these and the remaining factors were tested, and no significant results were obtained. To evaluate prognosis better in the presence of a homogeneous primary treatment, the 82 patients who presented at the study institution with primary disease and underwent surgery with curative intent were analyzed as a separate group.
Local Recurrence
FIGURE 3.
Five-year crude cumulative incidence of local recurrence by presentation status in the entire case series (167 patients). Solid line: primary tumor; dashed line: recurrent tumor.
62.7%) for the entire series. Patients with primary tumors had a 5-year crude cumulative incidence of local recurrences of 36.8% (95%CI, 25.1– 48.4%), whereas patients with recurrent tumors had a 5-year crude cumulative incidence of local recurrences of 71.5% (95%CI, 60.6 – 82.4%); the difference was highly significant (P ⬍ 0.01), as shown in Figure 3. In contrast, no significant difference was observed between patients with primary lesions and those with recurrent lesions with regard to the crude cumulative incidence of distant recurrences; in the entire series, the 5-year incidence of distant recurrences was 11.0% (95%CI, 5.9 –16.1%). The results obtained with the Cox model for overall survival are shown in Table 2. Statistical significance was achieved only for tumor grade (P ⫽ 0.02).
We selected tumor size, histology (liposarcoma vs. other histology), and tumor grade as possible prognostic factors in patients with primary disease only. We did not consider surgical margin status, because we judged that all surgical excisions of retroperitoneal sarcomas were marginal according to the Enneking criteria, even though they were macroscopically complete (regardless of whether a positive microscopic margin was found on pathologic examination). Significant results were obtained for tumor size (P ⫽ 0.04) and histology (P ⬍ 0.01). In particular, the 5-year crude cumulative incidence of local recurrence was 19.2% (95%CI, 2.3–36.2%) for tumors that measured ⱕ 10 cm in maximum dimension, 41.0% (95%CI, 20.2– 61.7%) for tumors that measured ⬎ 10 cm and ⱕ 20 cm in greatest dimension, and 47.8% (95%CI, 27.9 – 67.7%) for tumors that measured ⬎ 20 cm in greatest dimension. With regard to histology, the 5-year crude cumulative local recurrence rate was 57.7% (95%CI, 38.9 –76.5%) for patients with liposarcoma and 22.1% (95%CI, 9.2–35.1%) for patients with the other histotypes (Fig. 4). The 5-year crude cumulative local recurrence rate was nearly unaffected by tumor grade (P ⫽ 0.76) in that it was equal to 36.0% in patients with
2452
CANCER June 1, 2004 / Volume 100 / Number 11
FIGURE 4.
Five-year crude cumulative incidence of local recurrence by histology in primary cases (82 patients). Solid line: other histology; dashed line: liposarcoma.
FIGURE 5. Five-year crude cumulative incidence of distant metastasis by histology in primary cases (82 patients). Solid line: other histology; dashed line: liposarcoma.
Grade 1 tumors and 36.5% in patients with Grade 2–3 tumors.
failure of locoregional control. However, if patients with liposarcomas are separated from patients with nonliposarcomas, then the pattern of failure is different. In fact, liposarcomas were characterized mainly by a high local failure rate (Fig. 4), whereas distant metastases constituted the main cause of failure in patients with nonliposarcomas. Liposarcoma is the most frequent histotype by far among retroperitoneal sarcomas. This also was true in the current series, and the fact that the main cause of failure was local is a direct consequence of the prevalence of this histotype. It is interesting to note that the main cause of failure remains local, even when patients with low-grade liposarcomas and high-grade liposarcomas are separated. In fact, no metastatic spread was observed in any of the patients with liposarcoma who presented at the study institution with primary lesions. However, tumor grade does appear to correlate with prognosis, but mostly because it predicts the probability of local failure, and the cause of death is local invasion of vital structures. Even when areas of dedifferentiation are present within the tumor, the main problem of liposarcoma remains local control, and not distant metastases. The worst local control of liposarcoma, compared with other histotypes, may depend in part on tumor size at presentation: in the current series, liposarcomas had a median maximum tumor dimension of 28 cm (range, 8 – 65 cm) compared with 11 cm (range, 4 – 40 cm) for the other histotypes. This substantial difference may be related to differences in the pattern of growth, with a delayed onset of symptoms in patients who have liposarcoma compared with patients who have nonlipo-
Distant Metastases We attempted to determine whether the distant metastases parameter was affected by the same prognostic factors that were analyzed for local recurrence; however, the power of this analysis was limited by the relatively small number of events (10 metastases among patients with primary lesions). A significant result was obtained for tumor histology (P ⬍ 0.01) but not for tumor size (P ⫽ 0.36) or grade (P ⫽ 0.13). In particular, no distant recurrences were observed among patients with liposarcoma, and all 10 metastases occurred in patients with other histotypes, yielding a 5-year crude cumulative incidence of 24.9% (95%CI, 11.4 –38.4%) in this group (Fig. 5). It is interesting to note, despite a lack of statistical significance, that only 1 distant recurrence was observed among 28 patients with Grade 1 tumors, compared with 9 distant recurrences among 54 patients with Grade 2–3 lesions.
DISCUSSION In this series of 167 patients with retroperitoneal sarcomas who underwent surgery at a single institution over 20 years, the overall survival rate at 10 years after definitive surgery was 27%, and the disease-free survival rate was 16%. These results, although they were collected in a retrospectively reviewed, nonselected series, are superimposable to the main published experiences (Table 3). It is interesting to note that only a minority of patients developed distant metastases. Therefore, the relatively poor prognosis for patients with retroperitoneal sarcoma depends largely on the
Patterns of Recurrence in Retroperitoneal STS/Gronchi et al.
2453
TABLE 3 Complete Resection and Survival in Major Published Series Reference
Study period
No. of patients
Complete resection (%)
5-yr OS (%)
10-yr OS (%)
Lewis et al., 19982 Stoeckle et al., 20013 Ferrario and Karakousis, 20034 Dalton et al., 19897 Catton et al., 19948 Singer et al., 19955 Van Doorn et al, 19949 Kilkenny et al., 19966 Current study
1982–1997 1980–1994 1977–2001 1963–1982 1975–1988 1970–1994 1973–1990 1970–1994 1982–2001
231a 145a 130 116 104 83 70 63 167
80 65 95 54 43 NR 43 78 88
54 49 60 59 36 60 35 48 54
35 NR 48 NR 14 50 NR 37 27
OS: overall survival; NR: not reported. a Only patients with primary tumors.
sarcoma. In addition, nonliposarcomas may have a greater tendency to remain fairly circumscribed. Nevertheless, patients with nonliposarcomas were less numerous in the current series; therefore, any conclusion reached with regard to these patients should be regarded as preliminary. However, distant metastases clearly were an important cause of failure in this subgroup. These patients had mainly highgrade tumors. Of course, the relatively higher tendency of nonliposarcomas to metastasize may lead to the death of some of these patients before local recurrences can develop. However, masses that present in a more circumscribed fashion appear to have a lower tendency for locoregional spread. The risk of recurrence was found to be correlated with whether patients underwent surgery at the study institution for primary disease or recurrent disease: The disease-free survival rate was 27% at 10 years for patients with primary disease and 4.6% at 10 years for patients with recurrent disease. In principle, this result may depend either on the fact that the quality of initial surgery is crucial or on inherent prognostic factors that affect both the probability of local recurrence and the long-term outcome. Indeed, whether local recurrence depends mainly on tumor biology or on surgical adequacy has long been a matter of debate.2–10 In many series, the quality of surgery was the only significant prognostic factor for survival; therefore, extensive surgery has been advocated strongly.2–10 However, even in series that produced excellent rates of resection, the local recurrence rate remained high. This most likely is related to the difficulty of ensuring truly tumor free margins in retroperitoneal sarcomas, no matter how experienced the surgeon. However, at the least, the results of the current study do not exclude the finding that the adequacy of surgery is an important prognostic factor for patients with retroperitoneal sarcomas.
What practical conclusions may be drawn from the current study results regarding therapy and directions for future study? First, the fact that the quality of surgery is crucial for the treatment of a disease that recurs mainly locoregionally supports a policy of referring these patients to institutions with surgical expertise in the disease. The strongest reason, however, remains the rarity of the disease. Second, the histotype may make some difference. Liposarcoma clearly has a distinct pattern of growth (and therefore a different pattern of failure) compared with other histotypes. The main problem is that of local control, and tumor grade is a predictor of the time to local failure. Distant failures are far less of a problem. Conversely, nonliposarcomas are mainly high-grade tumors with a tendency for both local recurrence and distant metastases. Nonetheless, local control may be easier in selected tumors, with tumor grade representing mainly a predictor of distant failure, such as it does in adult STS at any other site. Therefore, it may be wondered what the roles of surgery, radiation therapy, and chemotherapy are in the treatment of patients with retroperitoneal sarcomas once they have been divided into liposarcoma and nonliposarcoma subgroups. Certainly, the need for adequate surgery should be stressed both for patients with liposarcoma and for patients with nonliposarcoma. However, although the procedure easily may miss its objective in the former group, it may be crucial in the latter group. Unfortunately, the role of radiation therapy, the adjuvant effectiveness of which has been demonstrated for patients with STS of the limb,21,27 is limited by the difficulty in delivering adequate doses to the retroperitoneum and the abdomen. In this regard, there are ongoing trials of preoperative radiation associated with a perioperative (intraoperative or postoperative) boost.11–20 This novel approach is interesting, but it remains experimental. With re-
2454
CANCER June 1, 2004 / Volume 100 / Number 11
gard to adjuvant chemotherapy, recent data point to its possible benefit in the treatment of adult STS, provided full doses are employed and high-risk patients are selected:22,23 If so, then patients with highgrade nonliposarcomas may benefit from chemotherapy for distant metastases. However, this remains to be demonstrated formally. Patients with high-grade liposarcomas may respond to chemotherapy; thus, a benefit from adjuvant chemotherapy with regard to local control cannot be ruled out in this subgroup. Therefore, by extrapolating from studies in other adult patients with STS, it is reasonable to believe that adjuvant chemotherapy may be useful in the treatment of patients with retroperitoneal high-grade sarcomas, especially with regard to distant control for nonliposarcomas and to local control for liposarcomas. The paucity of patients with retroperitoneal sarcoma makes it difficult to perform adjuvant studies targeted to them alone; therefore extrapolation may be a reasonable tool for medical decision-making. However, even if this premise is accepted, there remains a need for further confirmatory evidence of the effectiveness of adjuvant chemotherapy in the whole group of adult patients with STS. The results of the current series demonstrate that local recurrence after first surgery and highgrade malignancy are associated with poor survival. In addition, the histologic subtype strongly influences the pattern of recurrence, with liposarcomas recurring mainly locally and nonliposarcomas recurring in a predominantly distant pattern of failure. For this reason, and somewhat paradoxically, there may be a better chance for a permanent cure, however slim, in patients who have nonliposarcomas compared with patients who have liposarcomas. This should be taken into account when designing future clinical studies in patients with retroperitoneal soft tissue sarcomas. We will need prospective studies that assess the value of aggressive surgery, with liberal en bloc resection of adjacent organs, in patients with primary disease. This may be especially important for both patients with liposarcomas, who tend to develop locoregional recurrences, and for patients with nonliposarcomas, who have an even greater possibility of achieving local control. The role of the various forms of radiation therapy in this difficult anatomic region should be evaluated further. If it can be demonstrated conclusively that adjuvant chemotherapy has value in the treatment of soft tissue sarcoma at other sites, then its use in treating retroperitoneal sarcomas should be investigated further in terms of distant control in patients with high-grade, nonli-
posarcomas and also, possibly, in terms of local control in patients with low-grade liposarcomas.
REFERENCES 1.
2.
3.
4. 5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Brennan M, Alektiar KM, Maki RG. Soft tissue sarcoma. In: DeVita VT, Hellmann S, Rosenberg SA, editors. Cancer. Principles and practice of oncology. Philadelphia: Lippincott Williams & Wilkins, 2001:1841–1891. Lewis JJ, Leung D, Woodruff JM, et al. Retroperitoneal soft tissue sarcoma. Analysis of 500 patients treated and followed at a single institution. Ann Surg. 1998;228:355–365. Stoeckle E, Coindre JM, Bonvalot S, et al. Prognostic factors in retroperitoneal sarcoma. A multivariate analysis of a series of 165 patients of the French Cancer Center Federation Sarcoma Group. Cancer. 2001;92:359 –368. Ferrario T, Karakousis CP. Retroperitoneal sarcomas: grade and survival. Arch Surg. 2003;138:248 –251. Singer S, Corson JM, Demetri GD, et al. Prognostic factors predicting of survival for truncal and retroperitoneal soft tissue sarcoma. Ann Surg. 1995;221:185–195. Kilkenny JW, Bland KI, Copeland EM. Retroperitoneal sarcoma: the University of Florida experience. J Am Coll Surg. 1996;182:329 –339. Dalton RR, Donohue JH, Mucha P, et al. Management of retroperitoneal sarcomas. Surgery. 1989;106:725–733. Catton CN, O’Sullivan B, Kotwall C, et al. Outcome and prognosis in retroperitoneal soft tissue sarcoma. Int J Radiat Oncol Biol Phys. 1994;29:1005–1010. Van Doorn RC, Gallee MPW, Hart AAM, et al. Resectable retroperitoneal soft tissue sarcoma. The effect of extent of resection and post-operative radiation therapy on local tumor control. Cancer. 1994;73:637– 642. Heslin MJ, Lewis JJ, Nadler E, et al. Prognostic factors associated with long-term survival for retroperitoneal sarcoma: implications for management. J Clin Oncol. 1997;15:2832– 2839. Sindelar WF, Kinsella TJ, Chen PW, et al. Intraoperative radiotherapy in retroperitoneal sarcomas. Final results of a prospective, randomized, clinical trial. Arch Surg. 1993;128: 402– 410. Petersen IA, Haddock MG, Donohue JH, et al. Use of intraoperative electron beam radiation therapy in the management of retroperitoneal soft tissue sarcoma. Int J Radiat Oncol Biol Phys. 2002;52:469 – 475. Gieshen HL, Spiro IJ, Suit H, et al. Long-term results of intraoperative electron beam radiation therapy for primary and recurrent retroperitoneal soft tissue sarcoma. Int J Radiat Oncol Biol Phys. 2001;50:127–131. Bussieres E, Stockle EP, Richaud PM, et al. Retroperitoneal soft tissue sarcomas: a pilot study of intraoperative radiation therapy. J Surg Oncol. 1996;62:49 –56. Gilbeau L, Kantor G, Stoeckle E, et al. Surgical resection and radiotherapy for primary retroperitoneal soft tissue sarcoma. Radiother Oncol. 2002;65:137–143. Dubois JB, Debrigode C, Hay M, et al. Intraoperative radiotherapy in soft tissue sarcomas. Radiother Oncol. 1995;34: 160 –163. Alektiar KM, Hu K, Anderson L, et al. High dosed rate intraoperative radiation therapy (HDR-IORT) for retroperitoneal sarcomas. Int J Radiat Oncol Biol Phys. 2000;47:157–163. Jones JJ, Catton CN, O’Sullivan B, et al. Initial results of a trial of preoperative external-beam radiation therapy and postoperative brachytherapy for retroperitoneal sarcoma. Ann Surg Oncol. 2002;9:346 –354.
Patterns of Recurrence in Retroperitoneal STS/Gronchi et al. 19. Pisters PWT, Ballo MT, Patel SR. Preoperative chemoradiation treatment strategies for localized sarcoma. Ann Surg Oncol. 2002;9:535–542. 20. Sondak VK, Robertson JM, Sussman JJ, et al. Preoperative idoxuridine and radiation for large soft tissue sarcomas: clinical results with five-year follow-up. Ann Surg Oncol. 1998;5:106 –112. 21. Yang JC, Chang AE, Baker AR, et al. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcoma of the extremities. J Clin Oncol. 1998;16:197–203. 22. Tierney JF. Adjuvant chemotherapy for localised resectable soft-tissue sarcomas of adults: meta-analysis of individual data. Lancet. 1997;350:1647–1654. 23. Frustaci S, Gherlinzoni F, De Paoli A, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremity
24.
25.
26.
27.
2455
and girdles: results of the Italian Randomized Cooperative Trial. J Clin Oncol. 2001;19:1238 –1247. Trojani M, Contesso G, Coindre JM, et al. Soft-tissue sarcomas of adults; study of pathological prognostic variables and definition of a histopathological grading system. Int J Cancer. 1984;33:37– 42. Marubini E, Valsecchi MG. Analyzing survival data from clinical trials and observational studies. Chichester, U.K.: John Wiley & Sons Ltd., 1995. Gray RJ. A class of k-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141– 1154. Pisters PWT, Leung DHY, Woodruff J, et al. Analysis of prognostic factors in 1041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol. 1996;14: 1679 –1689.