Long-term recurrence of soft tissue sarcomas: Prognostic factors and implications for prolonged follow-up
See also editorial on pages 2942–3, this issue.
Abstract
BACKGROUND
To the authors' knowledge, the incidence of late recurrence (> 5 years after initial management) is unknown and no prognostic factors for late events have been characterized in patients with soft tissue sarcomas.
METHODS
Follow-up data from patients with localized soft tissue sarcoma who were included in the French Sarcoma Group database from January 1990 to June 2005 were reviewed. The outcomes of interest were the cumulative probabilities of late (> 5 years) local and metastatic disease recurrence with death as a competing event. Estimations and 95% confidence intervals (95% CIs) were computed with the cumulative incidence function.
RESULTS
A total of 719 patients who were alive and event free > 5 years after their initial diagnosis were included in the current study. Sixty-seven patients (9.3%) developed a late local recurrence and 42 patients (5.8%) developed a late metastatic recurrence, respectively. On multivariate analysis, internal trunk location (hazard ratio [HR], 3.9; 95% CI, 2.2-6.7 [P < .001]) and tumor size > 100 mm (HR, 2.1; 95% CI, 1.1-4 [P = .035]) were the 2 factors found to be independently associated with an increased risk of late local recurrence. Grade > 1 (graded according to the French Federation of Cancer Centers Sarcoma Group) (HR, 4.7; 95% CI 1.1-21 [P = .04]) was the sole factor found to be independently associated with an increased risk of late metastatic recurrence.
CONCLUSIONS
Late recurrence of soft tissue sarcoma is relatively uncommon. However, the results of the current study emphasize the critical role of long-term follow-up to detect late local disease recurrence in patients with retroperitoneal or very large soft tissue sarcomas, and late metastatic recurrence in patients with high-grade disease. Conversely, the prolonged follow-up of patients with grade 1 disease is not needed. Cancer 2014;120:3003–3006. © 2014 American Cancer Society.
INTRODUCTION
Soft tissue sarcoma (STS) is a rare and heterogeneous disease that comprises > 50 distinct histological subgroups.1 The treatment of STS in specialized centers is expressly recommended.2-4 However, to our knowledge, there is no consensus regarding the follow-up of patients with STS after initial management.5
Late disease recurrence is generally defined as recurrence that occurs > 5 years after initial management. The incidence of late recurrence, which would justify prolonged follow-up, is not well-documented and to our knowledge no prognostic factors specific for late events such as local recurrence or metastasis have been characterized in patients with STS. The diversity of STS in terms of anatomic location, histology, and biological behavior makes global analysis inappropriate. Early diagnosis of local recurrence or metastatic recurrence can lead to aggressive treatments with curative intent and prolonged survival in selected patients.6 Prognostic factors for a tailored follow-up are therefore needed.
The primary objective of the current study was to determine prognostic factors for late local and metastatic disease recurrence to better distinguish those patients who should have a longer and more careful follow-up from those who do not need it.
MATERIALS AND METHODS
The current study was based on data from the French Sarcoma Group database (conticabase.sarcomabcb.org/). Inclusion criteria were 1) a primary diagnosis of localized deep STS after January 1990; 2) macroscopically complete surgery; and 3) complete remission maintained for at least 5 years after initial management.
The outcomes of interest were the cumulative probabilities of late (> 5 years) local and metastatic disease recurrence, respectively, with death as a competing event. In all cases, the histological diagnosis was established according to the World Health Organization Classification of Tumors1 by an expert pathologist. The histological grade was determined as previously described according to the French Federation of Cancer Centers Sarcoma Group (FNCLCC) grading system.7
Statistical Analysis
Descriptive statistics were used to show the distribution of variables in the population. We used a proportional hazards model for the subdistribution of competing risks. Estimations and 95% confidence intervals (95%CIs) were computed with the cumulative incidence function. Patients who did not experience the event of interest or death over the course of the study were censored at the time of their last follow-up.
Variables tested in univariate analysis included age, sex, location, size, histology, grade, depth, bone/neurovascular involvement, surgical margins, adjuvant radiotherapy, and adjuvant chemotherapy. Variables associated with the respective endpoint of interest with a P value < .05 were planned to be included in the multivariate analysis.
RESULTS
A total of 3369 patients with STS underwent macroscopically complete surgery for a localized STS between 1990 and 2009. Among them, 719 patients were free of disease recurrence 5 years after the initial diagnosis and were therefore included in the current study. Their characteristics are described in Table 1.
| Characteristic | No. of Patients | % |
|---|---|---|
| Median age (range), y 52 (16-92) | ||
| Sex | ||
| Male | 352 | 49 |
| Female | 367 | 51 |
| Location | ||
| Limb | 509 | 70.8 |
| Trunk wall | 70 | 9.7 |
| Head and neck | 18 | 2.5 |
| Internal trunk | 122 | 17 |
| Median tumor size (range), mm 80 (10-600) | ||
| Histology | ||
| Liposarcoma | 233 | 32.4 |
| Well differentiated/dedifferentiated | 148 | 20.5 |
| Myxoid/round cell | 66 | 9.2 |
| Pleiomorphic | 19 | 2.6 |
| Undifferentiated sarcoma | 184 | 25.6 |
| Leiomyosarcoma | 81 | 11.3 |
| Synovial sarcoma | 72 | 10 |
| MPNST | 25 | 3.5 |
| Rhabdomyosarcoma | 20 | 2.8 |
| Fibrosarcoma | 18 | 2.5 |
| Other | 86 | 11.9 |
| FNCLCC grade | ||
| 1 | 189 | 26.3 |
| 2 | 265 | 36.9 |
| 3 | 244 | 33.9 |
| Unknown | 21 | 2.9 |
| Surgical margins | ||
| R0 | 326 | 45.3 |
| R1 | 129 | 17.9 |
| Unknown | 264 | 36.7 |
| Radiotherapy | ||
| No | 186 | 25.9 |
| Neoadjuvant | 3 | 0.4 |
| Adjuvant | 530 | 73.7 |
| Chemotherapy | ||
| No | 434 | 60.4 |
| Adjuvant | 177 | 24.6 |
| Neoadjuvant | 108 | 15.0 |
- Abbreviations: FNCLCC, French Federation of Cancer Centers Sarcoma Group; MPNST, malignant peripheral nerve sheath tumor.
The median follow-up was 9.1 years (95% CI, 8.7 years-9.8 years). At the time of analysis, 96 patients were dead of any cause. Sixty-seven patients (9.3%) developed a late local disease recurrence whereas 42 patients (5.8%) developed a late metastatic recurrence.
Prognostic Factors Associated With Late Local Recurrence
On univariate analysis, internal trunk location, a liposarcoma histological subtype, tumor size > 100 mm, R1 surgical margins, and lack of adjuvant radiotherapy were found to be significantly associated with an increased risk of late local recurrence (Table 2).
| No. of Patients | Events | % | HR (95% CI) | P | P | |
|---|---|---|---|---|---|---|
| Location | ||||||
| Limb | 509 | 29 | 5.7 | Reference | .708 | |
| Trunk wall/head and neck | 88 | 7 | 8.0 | 1.18 (0.50-2.77) | <.001 | <.0001 |
| Internal trunk | 122 | 31 | 25.4 | 4.98 (3.01-8.25) | ||
| Histology | ||||||
| Liposarcoma | 233 | 31 | 13.3 | Reference | .108 | |
| Leiomyosarcoma | 81 | 6 | 7.4 | 0.49 (0.20-1.17) | .045 | .0480 |
| Undifferentiated sarcoma | 184 | 14 | 7.6 | 0.52 (0.28-0.99) | .024 | |
| Other | 221 | 16 | 7.2 | 0.50 (0.27-0.91) | ||
| Tumor size, mm | ||||||
| <51 | 195 | 12 | 6.2 | Reference | .689 | |
| 51-101 | 246 | 13 | 5.3 | 0.85 (0.39-1.88) | .009 | .0006 |
| >100 | 249 | 34 | 13.7 | 2.43 (1.25-4.72) | ||
| Unknown | 29 | 8 | 27.6 | — | ||
| Surgical margins | ||||||
| R0 | 326 | 22 | 6.7 | Reference | .001 | .0005 |
| R1 | 129 | 21 | 16.3 | 2.85 (1.56-5.21) | ||
| Unknown | 264 | 24 | 9.1 | — | ||
| Radiotherapy | ||||||
| No | 186 | 25 | 13.4 | Reference | .004 | .0038 |
| Yes | 533 | 42 | 7.9 | 0.48 (0.29-0.79) |
- Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.
On multivariate analysis, internal trunk location (hazard ratio [HR], 3.9; 95% CI, 2.2-6.7 [P < .001]) and tumor size > 100 mm (HR, 2.1; 95% CI, 1.1-4 [P = .035]) were the 2 factors found to be independently associated with an increased risk of late local disease recurrence.
Prognostic Factors Associated With Late Metastatic Recurrence
On univariate analysis, leiomyosarcoma histological subtype and grade > 1 were significantly associated with an increased risk of late metastatic recurrence (Table 3).
| No. of Patients | Events | % | HR (95% CI) | P | P | |
|---|---|---|---|---|---|---|
| Histology | ||||||
| Liposarcoma | 233 | 9 | 3.9 | Reference | .016 | |
| Leiomyosarcoma | 81 | 10 | 12.3 | 3.06 (1.23-7.61) | .280 | .0027 |
| Undifferentiated sarcoma | 184 | 4 | 2.2 | 0.52 (0.16-1.69) | .059 | |
| Other | 221 | 19 | 8.6 | 2.14 (0.97-4.70) | ||
| FNCLCC grade | ||||||
| 1 | 189 | 3 | 1.6 | Reference | .021 | |
| 2 | 265 | 3 | 6.8 | 4.21 (1.24-14.3) | .016 | .0253 |
| 3 | 244 | 18 | 7.4 | 4.54 (1.33-15.5) | ||
| Unknown | 21 | 18 | 14.3 | — |
- Abbreviations: 95% CI, 95% confidence interval; FNCLCC, French Federation of Cancer Centers Sarcoma Group; HR, hazard ratio.
On multivariate analysis, grade > 1 (HR, 4.7; 95% CI, 1.1-21 [P = .04]) was found to be the sole factor independently associated with an increased risk of late metastatic disease recurrence.
DISCUSSION
The results of the current study demonstrate that late disease recurrence among patients with STS is relatively uncommon. By distinguishing local recurrence from metastatic recurrence in the current analysis, we found that each pattern of recurrence occurs preferentially in distinct subtypes of STS and that they are quite different in terms of prognostic factors.
Late local recurrence occurred in 25% of patients with internal trunk sarcoma and 14% of patients with a tumor size > 100 mm. Retroperitoneal sarcomas share both negative prognostic features for local recurrence. More specifically, retroperitoneal liposarcomas, especially when well differentiated, can have a very indolent course interspersed with iterative recurrences. Indeed, a recent work from our group demonstrated that the local recurrence-free survival of patients with retroperitoneal well-differentiated and dedifferentiated liposarcomas is divided by 2 between 5 years and 10 years.2 Therefore, a tailored long and regular follow-up appears justified for these patients.
Although the majority of metastatic recurrences occur within 2 years from the time of the initial diagnosis, the data from the current study indicate that 1 in 14 patients with high-grade STS who was free of recurrence at 5 years developed a late metastatic recurrence. One mechanism that could explain the late recurrence of high-grade, highly proliferating tumors is tumor dormancy. To the best of our knowledge, cancer dormancy is poorly understood, and various mechanisms have been reported, including angiogenic dormancy, cellular dormancy (G0-G1 arrest), and immunosurveillance.8 A large set of compelling preclinical data have suggested that the angiogenesis switch could be a crucial trigger in reversing tumor dormancy.9, 10 These data indicate the need for a deeper understanding of the underlying tumor biology driving dormancy in patients with high-grade sarcomas.
To the best of our knowledge, there are no published data to indicate the optimal routine follow-up policy for patients with localized STS who are treated surgically. Current recommendations suggest following surgically treated patients with intermediate-grade to high-grade disease every 3 to 4 months in the first 2 to 3 years, then twice yearly up to the fifth year and once a year thereafter.2 It is also proposed to follow patients with low-grade sarcomas for local recurrence every 4 to 6 months in the first 3 to 5 years and then annually thereafter. It is interesting to note that recent data have suggested no difference in time to detection and overall survival between less costly (eg, chest x-ray at 6-month intervals) and more expensive (computed tomography scans at 3-month intervals) testing or follow-up regimens.11, 12
The results of the current study emphasize the critical role of long-term follow-up to detect late local disease recurrence in patients with retroperitoneal or very large STS, and late metastatic recurrence in patients with high-grade disease. In addition to current European Society for Medical Oncology recommendations for the first 5 years, we suggest that patients with retroperitoneal sarcoma and patients with high-grade tumors be followed biannually for up to 10 years (eg, with biannual chest x-ray and annual computed tomography scans). However, the data from the current study also suggest that the prolonged follow-up of patients with grade 1 nonretroperitoneal disease is not needed.
FUNDING SUPPORT
Supported by the French National Cancer Institute (INCa) grant INCa-DGOS-Inserm 6046.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
