Comparative Analysis of Skip Metastasis in Pediatric Osteosarcoma: Clinical Features and Outcomes

Abstract

Background:

Skip metastasis (SM) is a synchronous regional bone metastasis. Using new imaging modalities, the detection of SM is easier and possibly more common. We reviewed patients with SM and compared their characteristics and outcomes to other patients with osteosarcoma treated at our center.

Methods:

We reviewed retrospectively children (<18 years) with newly diagnosed osteosarcoma who presented from June 2006 to March 2022. Patients’ characteristics, treatment modalities, and outcomes were analyzed. All cases were discussed in a multidisciplinary clinic that included 2 experienced radiologists.

Results:

We identified 155 patients with osteosarcoma, among which 13 (8.3%) patients had SM detected by MRI. Patients with SM had a median age at diagnosis of 11.2 years (range 7 to 17). Three patients had lung metastasis at diagnosis. Bone scan was positive for the SM in 8 patients (62%). All patients underwent primary tumor resection after neoadjuvant chemotherapy (amputation in 5, limb salvage surgery in 8). Five had postchemotherapy necrosis ≥90% in primary tumor. Seven patients relapsed/progressed (1 local and 6 in the lung), all relapsed patients died of disease. Compared to the rest of the patients, those with SM had similar clinical features to patients without SM; outcomes were similar with no significant differences in event-free survival and overall survival (P=0.7 and 0.3, respectively).

Conclusion:

In this study, we observed a percentage of patients with SM comparable to previous reports. Patients with SM exhibited clinical features akin to the rest of our patients. Thorough evaluation of imaging studies and multidisciplinary care, coupled with meticulous surgical planning, are crucial for achieving a cure, which remained unjeopardized in our patients with SM.

Enneking defined skip metastases (SM) as synchronous foci of bone metastases that are anatomically separate from the primary tumor in the absence of distant skeletal metastases. Typically, SM occurs at the same bone as the primary tumor, or they are seldom detected in the bone opposite the joint, in which case they are known as trans-articular SM.^1,2 The American Joint Committee on Cancer (AJCC) defined skip metastasis as “2 or more discontinuous lesions in the same bone”.³ The incidence of SM varied between studies, ranging from 1.4% to 25%.^4–8

Various imaging techniques, including bone scans, computed tomography (CT), and magnetic resonance imaging (MRI), have been employed to identify SM. Newer imaging modalities offer higher-resolution, potentially facilitating easier and more frequent detection of SM. Nonetheless, MRI remains the most sensitive technique.^9–12

The outcome of patients with osteosarcoma is affected by multiple factors, for example, metastasis, poor histologic response (necrosis less than 90%), and early disease progression.^13–17 Presence of distant metastasis; in particular bone metastasis, is reported to have the worst effect on survival.^15,16 The impact of SM on the prognosis of patients with osteosarcoma remains a topic of controversy.^{4,5,8,11,17,18} Various studies have presented conflicting findings; some report a poor outcome, whereas others show similar survival rates to patients without SM when patients receive appropriate treatment through a combination of surgery and chemotherapy.^7,8,17–19 The failure to remove the SM surgically has been identified as a potential factor contributing to local recurrence.^7,9,20,21

In this study conducted at a single center, we investigated the incidence, clinical features, and prognosis of SM in patients with osteosarcoma. Additionally, we conducted a comparative analysis of their outcomes with those of patients who did not have SM.

PATIENTS AND METHODS

Following approval from the institutional review board (IRB-23 KHCC 88), we conducted a retrospective review of medical records of pediatric patients 18 years of age or younger, newly diagnosed with osteosarcoma. The study included patients treated at our institution between June 2006 and March 2022. Inclusion criteria encompassed patients with localized extremity primary disease or isolated lung metastasis, while excluding those with multiple metastatic sites since they were treated with palliative intent due to the known dismal outcome, nonextremity primary disease, or those who only sought consultation or underwent surgery without further management.

We collected demographic and clinical data, including age, gender, results of MRI and chest CT at the time of diagnosis and before local control, local treatment approach, and percentage of necrosis in primary tumor following neoadjuvant chemotherapy. For SM, we looked at site(s) of metastases in relation to the primary tumor, radiologic response after neoadjuvant chemotherapy, and treatment of the lesion(s).

Good histologic response to chemotherapy was defined as necrosis ≥90%. Lung recurrence was defined as the appearance of new lung nodules or reappearance of 1 or more lung nodules after response in preexisting lung metastasis. The size of the tumor was defined as large if primary osteosarcomas affecting ≥1/3 of the affected bone. Disease progression was defined as an increase in the size of the primary mass by ≥20% or an increase in the size of existing lung nodules while on therapy. Outcomes in terms of recurrence/progression, sites of recurrence, and the duration of follow-up were captured. Follow-up data were pursued until June 2023.

The treatment regimen was according to the European and American Osteosarcoma study (EURAMOS-1) protocol, where patients were scheduled to undergo 2 cycles of MAP (methotrexate, doxorubicin, and cisplatin) before local control of the primary tumor. The preferred local control method was limb salvage surgery (LSS) whenever feasible, with a focus on achieving negative resection margins. In cases where LSS was not achievable, alternative control was by amputation. If surgery was declined by the patient/family, then radiation therapy was considered. Patients with SM that persist after neoadjuvant chemotherapy were planned for lesion resection at the time of local control.

Every case was thoroughly reviewed during both the initial presentation and the local control phase at the multidisciplinary clinic (MDC). The MDC included a team of medical professionals, including a pediatric oncologist, orthopedic surgeon, pathologist, radiologist, and radiation oncologist, ensuring comprehensive discussions and evaluations for each case.

Statistical Analysis

Demographic, tumor, and treatment characteristics were summarized by descriptive analysis. Overall survival (OS) was defined as the time between diagnosis and death from any cause or last follow-up for patients remaining alive. Event-free survival (EFS) was defined as the time between diagnosis and occurrence of disease recurrence or progression, second malignancy, death, or last follow-up for patients who did not experience an event. OS and EFS estimates were calculated using the Kaplan–Meier method. Log-rank test was used to compare survival curves when needed. A P value of 0.05 or less was considered statistically significant.

RESULTS

During the course of our study, we were able to identify 155 patients who had received a diagnosis and treatment for osteosarcoma at our institution. Among this group, SM was detected in 13 patients, representing 8.3% of the total. To ensure a more focused analysis, we excluded 15 patients due to the presence of diffuse bone metastasis (metastatic bone lesions in places other than primary bone or trans-articular from primary), and an additional 6 patients with primary tumors located outside the extremities. As a result, our analysis ultimately included 134 patients.

Patients Characteristic

Patients With Skip Metastasis

Thirteen patients had SM identified by radiology at the time of diagnosis. The median age of diagnosis was 11.2 years (range, 7 to 17.3), 5 (38%) were males, and 3 patients (23%) had lung metastasis. All patients except 1 had their tumors in the lower extremities. Tumors were considered large, as the median diameter was 13.4 cm (range, 6.3 to 23.5) and 54% were ≥1/3 of the bone. Characteristics of those patients are listed in Table 1.

TABLE 1.

Patients’ Characteristics

Variable	Patients without skip (N=121), n (%)	Patients with skip (N=13), n (%)	P
Gender
Female	60 (50)	8 (62)	0.08
Male	61 (50)	5 (38)
Age, y
Median	12.9	11.2	0.4
Range	6.8–18	7–17
Tumor site
Upper extremity	12	1	>0.9
Lower extremity	109	12
Surgery
Amputation	27 (22)	5 (38)	0.4
LSS	83 (69)	8 (62)
No	11 (9)	0
Lung metastasis
Yes	32 (26)	3 (23)	>0.9
No	89 (74)	10 (77)
Necrosis
≥90%	51 (46)	5 (38)	0.5
<90%	59 (54)	8 (62)
Tumor diameter, cm			0.2
Median	10.8	13.4
Range	5–21.3	6.3–23.5
Size of primary
Small (<1/3)	56 (46)	5 (38)	0.4
Large (≥1/3)	46 (38)	7 (54)
NA	19 (16)	1 (8)

LSS indicates limb salvage surgery; NA, not available.

SM was detected by MRI in all patients, and by bone scan in 8 of 13 patients (62%). The number of SM was 1 in 8 patients, 2 lesions in 3, and multiple in the remaining 2. The site of SM was as a discontinuous lesion(s) at the same bone in 11 patients, whereas it was trans-articular in 2.

Treatment

All patients underwent treatment with neoadjuvant chemotherapy. Subsequent MRI scans indicated the continued presence of the skip lesion(s) in 11 patients, whereas in 2 patients the lesion(s) had completely disappeared.

Surgery as a local control was performed in all patients: LSS (n=8) or amputation (n=5). The persistent skip lesions were resected with primary tumors in 9 of the 11 patients. Postchemotherapy pathology showed good necrosis in the primary tumor in 5 patients. Information about the pathology of skip metastases was lacking except for 1 patient, where skip metastasis was confirmed by pathology to be necrotic.

Overall and Event-free Survival

The median follow-up time was 3.6 years (range, 0.7 to 15.1). All survivors were in their first complete remission. The 5-year EFS and OS were 50% (95% CI, 28% to 90%) and 84% (95% CI, 66% to 100%), respectively.

At the last follow-up, 7 patients developed events: lung recurrence (n=5), progression (n=1), or local recurrence (n=1) at a median of 2.2 years (range, 0.5 to 5.6) after initial diagnosis. All patients with events died of disease.

Comparison of Patients With and Without Skip Metastases

One hundred twenty-one patients did not have evidence of SM by images. Sixty-one (50%) were males with a median age at diagnosis of 12.9 years (range, 6.8 to 18), and 32 (26%) had lung metastasis at presentation. Characteristics of those patients are listed in Table 1. The 5-year EFS and OS were 51% (95% CI, 42% to 62%) and 64% (95% CI, 56% to 74%), respectively.

In comparison to patients with SM, there were no discernible differences in patient characteristics regarding age, gender, the presence of lung metastasis, tumor diameter, the type of surgery performed, or the extent of necrosis in the primary tumor (Table 1). There was no statistically significant difference in the 5-year EFS and OS between the 2 groups, with P values of 0.7 and 0.3, respectively (Fig. 1). The 5-year EFS and OS for patients who presented with SM but without lung metastasis were 55% (95% CI, 30% to 100%), which was comparable to that observed in patients without SM and without lung metastases (P=0.5), Fig. 2.

FIGURE 1.

Overall survival (OS) and event-free survival (EFS) for patients with and without skip metastasis.

FIGURE 2.

Overall survival (OS) and event-free survival (EFS) in a subgroup of patients with no lung metastasis, stratified by the presence or absence of skip metastasis.

DISCUSSION

Patients with osteosarcoma presenting with metastases, particularly bone metastases, have a poor outcome.^16,22 SM, which is a synchronous regional bone metastasis in osteosarcoma, and its effect on survival presents an important topic for investigation, especially with the advancement in imaging modalities resolution. The impact of SM on the overall outcome is not clear.^4–7,18 In this retrospective study, we did a comparative analysis between pediatric patients with osteosarcoma who presented with SM at the time of diagnosis and those who did not. We analyzed and compared the characteristics and the outcomes between the 2 groups.

The incidence of SM in patients with osteosarcoma varied widely in the literature, ranging from 1.4% to 25%;^4,6,10,18 this variation could be related to the time of the report, as the modalities of treatment changed over time, another possible reason is the method used for defining SM, whether it was based on imaging findings or pathology confirmation. In addition, some studies used Enneking definition that included trans-articular SM, whereas others included the AJCC definition that included SM in the same bone only. In our report, SM was identified in 8.3% of our study cohort; this included SM in the same bone or trans-articular, which aligns with the most recent studies in the literature. This consistency in findings among studies reinforces the significance of considering the presence of SM at the time of the diagnosis of osteosarcoma. Furthermore, our results highlight the importance of thorough staging and imaging techniques in the evaluation of pediatric patients with osteosarcoma.

The rate in our cohort is higher than that reported by the Cooperative Osteosarcoma Study Group (COSS).⁴ This difference may be attributed to the advanced disease status at the time of presentation in our study group, manifested with a large tumor diameter at presentation, as well as the detection methods employed. The COSS primarily included patients who were surgically diagnosed or identified with MRI progression only, potentially leading to underreporting of SM cases. Our cohort comprised patients diagnosed based on MRI results, which also may result in overdiagnosis of SM as some of these lesions may be benign.

Our findings suggest that, from a demographic and clinical standpoint, patients with SM closely resemble those without this metastatic pattern, which aligns with previous reports.^4,8,17 Importantly, the lack of significant disparities extended to the 5-year EFS and OS rates between the 2 groups. The 5-year EFS and OS rates for patients with SM were not significantly different from those without SM, as indicated by P values of 0.7 and 0.3, respectively. These findings suggest that the presence of SM at the initial diagnosis of osteosarcoma does not inherently confer a worse prognosis when compared to those without SM. This observation challenges previous notions that SM may be associated with more aggressive disease and underscores the complex and multifaceted nature of osteosarcoma.^2,8,11,17,18 Nevertheless, the limited sample size within our cohort may have led to a lack of statistical significance. Therefore, it is imperative to validate these findings in a larger patient cohort.

It is suggested that performing resections of both the metastatic disease including SM and the primary tumor is crucial to prevent local recurrence within the original bone.^4,21 In cases where the presence of SM could influence the surgical decision, it is advisable to conduct a biopsy to confirm the existence of malignant tissue.^4,7,9,10 In our patient cohort, biopsies were not routinely conducted. Instead, we primarily relied on MRI and bone scan results for SM diagnosis because, for the majority of patients, the decision to proceed with surgery was not affected by SM. Interestingly, 2 patients whose decision of surgery was affected by the site of SM, faced challenges during biopsy attempts, resulting in the SM being left untreated. Despite this, both patients are in their first remission 3.6 and 10 years since their initial diagnosis. It is important to acknowledge the potential for false-positive SM diagnoses when solely relying on MRI findings. In some reports, pathologic examination from some of those suspicious lesions revealed focal nodular marrow hyperplasia; this could explain the good outcome in those 2 patients.^4,20,23 In our cohort, the only patient who developed local recurrence had her lesion resected with the primary tumor, and the pathology confirmed necrosis of SM; therefore, the local recurrence was likely not related to the SM.

Bone scan did not dependably detect SM, as a negative result was detected in 5 of the 13 patients with MRI-positive SM, again this highlights the potential for false-positive SM diagnoses when solely relying on MRI findings. In 5 patients, the SM was in more than 1 focus, 3 of them died of disease. Whether this indicates that those patients with multiple SM have a more aggressive disease is not clear. PET/CT scan is emerging in the diagnosis, staging, and evaluation of treatment response in osteosarcoma.^24–27 Several studies have reported the superior accuracy of PET/CT in detecting bone metastasis when compared to CT scans, dedicated MRIs, and bone scan in bone sarcomas.^27–30 The exploration of the role of PET scans in assessing SM in osteosarcoma is an area that requires further investigation.

The current study has several limitations. The retrospective nature of the study and the long period of follow-up may have resulted in incomplete data for some patients. The current report is a single-institutional data, which may not represent the wider population. Additionally, the relatively small sample size of patients with SM warrants further investigation with a larger number of patients; this will help better understand the clinical implications and prognostic factors associated with SM. Another limitation of the study is the defining SM based on MRI findings without pathology confirmation, which may have led to overdiagnosis.

In conclusion, our study identified SM in 8.3% of pediatric patients with osteosarcoma; this highlights the need for attentiveness to this entity during imaging evaluation at diagnosis. In addition, our study showed similar clinical characteristics and outcomes of pediatric osteosarcoma patients with and without SM. It is important to note that a small sample size could potentially lead to an overestimation of these similarities. Our study highlights the importance of multidisciplinary care and optimal surgical planning in curing patients with osteosarcoma, particularly those with locally advanced lesions and those with SM.