Radiation-associated sarcoma of the pelvis and/or sacrum (RASB) is a rare but challenging disease process associated with a poor prognosis. POPS and RASB involving the pelvis and sacrum present challenging disease processes and their oncologic outcomes are similarly poor. However, the data presented here shows that perioperative and disease specific survival for patients with RASB is even worse than for patients with POPS. While surgery can result in a favorable curative outcome for a small subset of patients, surgical treatment is fraught with complications. As such, careful counseling is necessary to reach a patient-centered decision regarding the suitability and feasibility of proceeding with surgical treatment of these tumors.
1. Background
Radiotherapy is frequently utilized as part of curative cancer treatment, either as an adjuvant or with primary curative intent
[1]. Despite its widespread use, radiotherapy is associated with short- and long-term toxicity. In the short to mid-term, patients may develop fibrosis, delayed wound healing, and other complications
[2][3][4]. In the long term, more serious sequelae may manifest, such as radiation-associated malignancies
[1][5][6][7][8][9][10][11][12][13].
Radiation-associated bone sarcomas (RASB) are one such potential secondary malignancy. Prior studies demonstrated that when compared to primary bone sarcomas, perioperative and survival outcomes of RASB are worse
[7][8][11]. The pelvis is a common location for RASB due to the frequent use of radiotherapy as part of the treatment for genitourinary, gastrointestinal, and gynecological cancers. Standard treatment for pelvic bone sarcomas includes (neo)adjuvant chemotherapy and surgical resection with or without additional radiotherapy. Surgical options include internal or external hemipelvectomy and, depending on the location and extent of the soft tissue mass, additional vascular, visceral, or soft tissue reconstructive procedures may be required. As such, the decision to pursue curative treatment of a pelvic bone sarcoma requires careful, patient-specific multidisciplinary consideration as the procedures are associated with high morbidity, peri-operative mortality, and long recovery periods
[14][15][16]. Patients often require additional surgical procedures to manage post-operative complications, and once discharged from hospital they commonly need extended stays in rehabilitation facilities before being able to return home. Furthermore, patients may be left with chronic pain and/or permanent functional impairments, affecting mobility, bowel, and bladder function, significantly impacting their quality of life
[17][18]. As patients with RASB are known to have poor overall survival, the question of whether to offer these patients surgical resection is unclear. Previous small series which examined prognosis of patients showed that patients with RASB frequently present with advanced stages of disease, and even with surgical management survival is poor
[8][11].
2. Development and Findings
Radiation-associated sarcomas of the bony pelvis and sacrum are very challenging to treat and are often associated with a poor prognosis. Data to guide decision-making for these patients is limited. The addition of chemotherapy may mitigate this survival difference. Local recurrence or distant metastasis usually occurred within 2 years of surgery. Although surgery offers a chance for a cure in this patient population, post-operative complications are high.
Surgical resection in a previously irradiated field is technically challenging, often necessitating dissection through altered anatomy, fibrosis, and friable neurovascular structures. As such, it has been suggested that the need for amputation to manage these types of secondary malignancies may be higher
[7][10][19][20]. However, the rates of limb salvage did not differ between patients with RASB and POPS, and similar rates of negative margins were achieved. Difficulty in obtaining negative margins has been suggested as one possible reason for poor survival in patients with radiation-associated sarcomas involving bone and soft tissue. A study by Gladdy et al. investigated 130 patients with primary radiation-associated soft tissue sarcoma and found that margin status was an important independent predictor of survival
[1], findings that have been recapitulated by others
[13]. When considering patients with post-irradiation sarcoma of soft tissue and bone, similar findings have been suggested. Inoue et al. studied 61 patients with radiation-associated bone and soft tissue sarcomas who underwent surgical treatment and found that a wide surgical margin correlated with improved survival
[7]. This same relationship has also been suggested when considering only bone sarcomas. Kalra et al. investigated 42 patients with radiation-associated sarcomas of bone
[8] and found that complete surgical resection was the only independent determinant for survival.
While negative margin surgical resection is traditionally accepted as the mainstay of curative treatment for patients with pelvic and sacral bone sarcomas, (neo) adjuvant chemotherapy, especially for primary osteosarcoma and spindle cell sarcomas, provides additional disease control. However, the consideration of adjuvant radiotherapy or chemotherapy in the context of RASB is less well understood. Indeed, one of the challenges surrounding RASB is that treatment options are often more limited as patients are older and may have received prior chemotherapy, limiting subsequent options. Shaheen et al. showed that patients with a radiation induced sarcoma of bone of the extremities and pelvis demonstrated improved survival with combined treatment with both chemotherapy and surgery as compared to surgery alone
[11]. Bacci et al. similarly found that the use of chemotherapy in RASB resulted in survival rates that were not dissimilar from patients with primary conventional high-grade osteosarcoma
[21][22]. However, treatment regimens presented in these studies were heterogenous and these studies did not specifically account for appendicular vs. axial location when investigating the utility of adjuvant therapy. Patients with RASB were less likely to receive chemotherapy (70.3% received for RASB vs. 90.6% for POPS), though there was no difference in utilization of radiation between the two groups. When considering the individual groups, the addition of chemotherapy was associated with improved survival for patients with POPS (
p = 0.03) but not RASB (
p = 0.06). The lack of an association between chemotherapy utilization and improved DSS in RASB likely relates to the small size of that group of patients, as no difference in DSS when comparing RASB to POPS when controlling for the use of chemotherapy. Most patients receiving chemotherapy for a RASB received dual agent chemotherapy regimens that did not include methotrexate and often did not receive a full course of chemotherapy. There was no difference in “good” histological response following neoadjuvant chemotherapy between RASB and POPS. Failure to complete chemotherapy has been suggested as a risk for poorer survival
[23] and may be a contributing factor to the poorer DSS observed. While rates of methotrexate utilization differed, the significance of this finding is less clear. Several randomized trials failed to show a survival benefit with the addition of methotrexate for adult patients compared to dual agent regimens with Adriamycin and cisplatin alone
[24][25].
It is important to consider the morbidity and mortality associated with the surgical treatment of RASB when counseling patients. Surgery can result in favorable 5-year survival for ~25% of patients; however, it is critical to acknowledge that surgical treatment is fraught with morbidity, with as many as 80% of patients experiencing post-surgical complications and 15% dying in the perioperative period. Local recurrences and metastases occur typically occur within the first year post-operatively, with a median time to local recurrence and metastasis of 5 months and 8 months, respectively. Median survival was only 11 months and disease-free survival beyond 3 years was achieved in only 20% of patients. Given the extensive morbidity and extended recovery required following internal or external hemipelvectomy and sacrectomy
[14][15][16], the decision to proceed with surgery requires careful consideration.
The observation of a survival difference between RASB and POPS is more challenging to elucidate. Interestingly, although patients with a RASB were older and had less intensive chemotherapy treatment, similar oncologic outcomes were achieved overall. When controlling for chemotherapy utilization, DSS was indeed similar between groups, though it is possible that this subanalysis was underpowered to detect such a difference. This points potentially to a comparable tumor biology. In further support of this consideration, margin status was not different between groups and was not associated with DSS. Given that there was no difference in LRFS, then, again, there may not be an explicit difference in the biological aggressiveness of the two groups. This is surprising as there was no difference in LRFS or MFS between groups, so local recurrence and metastasis alone are unlikely to explain this difference. One possible explanation is the large difference in perioperative mortality between groups. Indeed, when excluding patients who died perioperatively, there was no observed difference in DSS between groups, though, again, this subanalysis may be underpowered to detect such a difference. On the other hand, though, it is possible that the high rates of perioperative complications and increased rates of perioperative mortality for RASB patients as compared to POP patients may be related to the challenges of operating after previous surgeries and/or after previous radiation. Thus, while the DSS difference observed may not be explicitly related to tumor biology, it may be explained by the challenges of operating after prior oncologic treatment in the same field.
It is reasonable to expect that complications may be higher and outcomes worse in this excluded palliative group and that taken as a whole, the results presented here may be more favorable than if all patients with RASB and POPS of the pelvis and sacrum were included. Furthermore, the exclusion of patients treated without definitive surgery prevents us from understanding the natural history of this disease and weighing the risks and benefits of surgical versus non-operative treatment. Future studies’ investigation RASB of the pelvis and sacrum should consider including patients undergoing palliative treatment. Another limitation is that data regarding functional outcomes was not available for many patients. This was likely because many of the patients did not survive long enough for reasonable functional assessments to be undertaken.