Keywords
clavicle - Ewing’s sarcoma - total clavicle excision - partial clavicle excision
Ashish Gulia
Introduction
Primary clavicle tumors are uncommon and account for <1% of primary bone tumors. Limited
case reports or case series are currently available that provide information on these
tumors.[1]
[2] The majority of primary clavicle tumors are malignant[3]
[4] and surgical treatment entails partial or total clavicle excision based on the disease
extent.[5]
Approximately 1.4% of all cases of Ewing’s sarcoma occur in the clavicle.[6] While chemotherapy is mandatory for cure, surgical excision is the preferred modality
for local control in Ewing’s sarcoma. Radiotherapy in addition to surgery or radiotherapy
alone is an acceptable option.
We report here a single institution study of Ewing’s sarcoma of clavicle treated with
surgical excision. We evaluated the functional and oncology outcomes and compared
outcomes between total and partial clavicle excisions.
Materials and Methods
Patients
This retrospective analysis included data from our prospectively maintained database.
Cases (N = 21) of histologically proven Ewing’s sarcoma of clavicle who underwent surgical
excision (partial or total clavicle excision) between January 2002 and December 2017
were included. The mean age of the patients was 15.3 (range: 6–40) years and the majority
(58%, 12/21) were males. [Table 1] presents the baseline demographics and characteristics of the patients. Baseline
imaging was done in the form of a plain radiograph and magnetic resonance imaging
of the shoulder region screening the entire clavicle.
Table 1
Patient demographics and baseline characteristics
|
Parameters
|
N = 21
|
|
Abbreviations: n, number of patients; SD, standard deviation.
|
|
Age (years), mean, range
|
16 (6–40)
|
|
Sex, n (%)
|
|
Male
|
12 (58%)
|
|
Female
|
9 (42%)
|
|
Details of surgery, n (%)
|
|
Total clavicle excision
|
13 (62%)
|
|
Partial clavicle excision
|
8 (38%)
|
|
Adjuvant treatment, n (%)
|
|
|
Chemotherapy plus radiotherapy
|
15 (71.5%)
|
|
Chemotherapy only
|
6 (28.5%)
|
Histopathology was confirmed with a core needle biopsy or histopathology review of
slides/blocks if a biopsy was done elsewhere. Staging investigations included computerized
tomography (CT) of the chest and either a fluorodeoxyglucose—positron emission tomography
(FDG-PET) scan or a combination of a bone scan and a bone marrow aspiration biopsy.
The final treatment plan was decided in multidisciplinary joint clinics (MDJC). Patients
with disseminated disease were not offered surgical excision and were treated with
palliative intent. These patients are not included in this analysis. Three (14%) of
the 21 patients in this series were metastatic at presentation (two—pulmonary and
one—lymph nodes). All the patients included in this study were treated with curative
intent. Two (10%) of the 21 patients had a pathological fracture through the clavicle
at presentation.
Chemotherapy
Patients were administered chemotherapy as per standard institute protocol. The protocol
included two courses of VIE couplet (vincristine, ifosfamide, and etoposide) followed
by two courses of VAC couplet (vincristine, doxorubicin, and cyclophosphamide) administered
every 3 weeks. Maintenance therapy after treatment of the primary tumor included 10
courses of alternating VAC and VIE couplets, with actinomycin D substituted for doxorubicin
after a total dose of 360 mg/m2. Vincristine was administered weekly throughout between the chemotherapy pulses.
Pathology
All resected specimens were evaluated for margin status and percentage necrosis.[7] Based on the percentage necrosis, patients were divided into good necrosis (necrosis
>90%) and poor necrosis (necrosis <90%).
Radiotherapy
Radiotherapy was usually recommended for those with poor response to chemotherapy,
in patients with a large soft tissue component at presentation and those who had an
open biopsy or pathological fracture. The target volume for radiation therapy included
the pretreatment extent of disease with appropriate margins. External beam radiotherapy
was delivered postoperatively to a dose of 45Gy/ 25#/5 weeks with conformal portals.
Follow-up
Patients were followed-up every 6 months for the first 5 years and annually till 10
years. Chest and local radiographs were done as part of imaging. Based on clinical
and radiological suspicion of local or distant recurrence, additional studies including
biopsy/fine needle aspiration cytology were done. Functional analysis using Musculoskeletal
Tumor Society (MSTS) scoring system was done at the time of the last follow-up. The
MSTS for the upper limb is based on the analysis of six factors, i.e., pain, function,
emotional acceptance, hand positioning, manual dexterity, and lifting ability. For
each of the six factors, values of 0 to 5 were assigned based on established criteria.
The result is expressed as a sum total with a maximum score of 30.[8]
Study Assessments
A detailed patient history in addition to gender, symptoms on presentation, histology,
treatment details in the form of surgery (partial/total clavicle excision), chemotherapy,
radiotherapy, local and distant recurrences, complications, current status, and the
MSTS score were recorded and compared. The study end points included the evaluation
of disease outcomes (locoregional recurrence free survival [LRFS], disease-free survival
[DFS], overall survival [OS] at 3 and 5 years) and functional MSTS scores.
Statistical Analysis
Demographic and baseline characteristics were summarized using descriptive statistics.
Categorical variables were summarized with frequency and percentage. Continuous variables
were summarized with count, mean, standard deviation, median, minimum and maximum.
Survival analysis was performed to measure lifetime or the length of time until the
occurrence of an event (disease recurrence in cases of LRFS, disease progression in
case of DFS, and death in case of OS). Survival data was analyzed using SPSS version
21 to measure the duration of time until a specified event occurs. LRFS, DFS, and
OS were calculated and analyzed using the Kaplan-Meier method and log-rank test.
Results
Surgery Details
Of 21 patients in this study, 60% (n = 13) patients had a total clavicle excision and 40% (n = 8) had a partial clavicle excision. A margin of 2 cm beyond disease or disarticulation
of the adjacent joint, if necessary, was considered adequate oncologic resection while
performing a partial clavicle excision.[9] No reconstruction was done in any case. One patient had an intraoperative rent in
the subclavian vein which was repaired. Wound necrosis was observed in one patient
which was managed conservatively. None of the patients had any permanent sequelae
of complications. Margins were negative in all cases.
Neoadjuvant Therapy
All patients received neoadjuvant chemotherapy followed by surgery between 10 and
12 weeks after starting chemotherapy. Fourteen patients (67%) had good response while
six (28%) patients had a poor response to chemotherapy; details were not available
in one patient. After MDJC discussion, radiotherapy was administered in 15 (71%) cases
(1—preoperative and 14—postoperative) in view of poor necrosis, large disease volume
at presentation, pathological fracture, or prior intervention in form of open biopsy
elsewhere. Of the two patients with a pathological fracture, one received postoperative
radiotherapy while the other patient was 11 years old with 100% necrosis for whom
MDJC decided not to administer radiotherapy. The patient is currently disease free
after 116 months. All the six patients who did not receive radiotherapy had good necrosis
and negative surgical margins of resection.
Survival
Of 21 patients, 20 had complete follow-up details. 10 patients died (nine due to disease
and one due to other reasons). Eight patients remain disease free and alive. Isolated
distance recurrence was observed in 10 patients of which two had lung metastasis at
presentation and one had nodal disease. Only two out of these 10 patients are alive,
one who underwent metastasectomy and is currently free of disease, and another patient
who is on palliative treatment (had nodal metastasis at presentation).
One local recurrence (5%) occurred at 12 months after completion of the treatment.
The local recurrence was resected again but eventually the patient died due to distant
metastasis.
The median follow-up of the entire cohort was 42 months (range: 7–198 months). In
survivors, the median follow-up was 92 months (range: 30–198 months). The LRFS, DFS,
and OS were 95, 59, 65% at 3 years, and 95, 47, 59% at 5 years, respectively. Poor
necrosis (<90%) had a significant impact on OS (p = 0.024). The administration of radiotherapy did not have a significant impact on
LRFS, DFS, or OS ([Table 2]).
Table 2
Response to chemotherapy and survival
|
5-y LRFS
|
5-y DFS
|
5-y OS
|
|
Abbreviations: DFS, disease-free survival; LRFS, loco-regional free survival; OS,
overall survival.
|
|
Good necrosis (>90%)
|
96%
|
51%
|
66%
|
|
Poor necrosis (<90%)
|
100%
|
25%
|
33%
|
|
p-Value
|
0.535
|
0.182
|
0.024
|
Shoulder Function
The MSTS was available for nine patients who are currently free of disease. The median
MSTS score was 29 (range: 27–30). The MSTS score after total clavicle excision ([Fig. 1]) was 28.5 (n = 5) compared with 29.5 (n = 4) for those who underwent partial clavicle excision ([Fig. 2]) (p = 0.125).
Fig. 1 (A) Radiograph at presentation, showing destructive lytic lesion of left clavicle with
large soft tissue component in a 6-year-old female diagnosed with Ewing sarcoma. (B) Post-chemotherapy radiographs showing good response to chemotherapy (reduction in
soft tissue component and union of pathological fracture). (C) Four-year follow-up radiograph post-total clavicle excision. (D–G) Excellent functional outcome.
Fig. 2 (A) Radiograph at presentation, showing lytic lesion of the lateral end of right clavicle
in a 16-year-old female diagnosed with Ewing sarcoma. (B) Post-chemotherapy radiographs showing good response to chemotherapy. (C) A 6-year follow-up radiograph post partial clavicle excision. (D–G) Excellent functional outcome.
Discussion
Primary clavicle tumors are rare accounting for <1% of bone tumors. In an analysis
by Klein, clavicle tumors were only 0.45% of more than 13,000 primary bone tumors.[10]
[11] The most common histologic diagnosis of clavicle tumors are myeloma, Ewing’s sarcoma,
and osteosarcoma.[12] As in other primary bone sarcomas, in the clavicle too, adequate resection of the
bone (partial or total clavicle excision) is the procedure of choice for best local
control.[13]
In a study published by Mayilvahan of six cases of Ewing’s sarcoma of the clavicle,
two died post-surgery due to disease.[11] Of the five cases published by Martin et al, one patient died of metastases post-surgery.[14] Our series of 21 cases with much larger numbers had a LRFS, DFS, and OS of 95, 59,
65% ([Table 4]) at 3 years and 95, 47, 59% at 5 years, respectively ([Table 3]).
Table 3
Radiotherapy and survival
|
Radiotherapy
|
5-y LRFS
|
5-y DFS
|
5-y OS
|
|
Abbreviations: DFS, disease-free survival; LRFS, loco-regional free survival; OS,
overall survival; RT, radiotherapy.
|
|
Yes
|
100%
|
32%
|
49%
|
|
No
|
83%
|
83%
|
83%
|
|
p-Value
|
0.141
|
0.065
|
0.110
|
Table 4
Comparison with published literature related to Ewing’s sarcoma of clavicle
|
Author
|
Total cases
|
Ewing’s sarcoma of clavicle
|
Follow–up (months)
(median)
|
Alive/Dead
(Only Ewing’s sarcoma)
|
Total/Partial clavicle excision
|
Mean functional outcome (number of cases with no reconstruction)
|
Mean functional outcome (number of cases with reconstruction)
|
5-y
LRFS
|
5-y
OS
|
|
Abbreviations: LRFS, loco-regional free survival; MSTS, Musculoskeletal Tumor Society;
OS, overall survival; RT, radiotherapy.
|
|
Natarajan et al[11]
|
12
|
6
|
48
|
4/2
|
7/5
|
5-Excellent
7-Good (12)
(MSTS)
|
NA
|
NA
|
NA
|
|
Li et al[12]
|
11
|
3
|
38
|
3/0
|
5/6
|
27.5 (5)
(MSTS)
|
28.8 (6)
(MSTS)
|
NA
|
NA
|
|
Rodriguez Martin et al[14]
|
5
|
5
|
12
|
4/1
|
1/4
|
93.5 (2)
(Constant shoulder score)
|
77.5 (2)
(Constant shoulder score)
|
NA
|
NA
|
|
Our study
|
21
|
21
|
42
|
10/10
|
13/8
|
29 (9)
(MSTS)
|
NA
|
95%
|
59%
|
Even Ewing’s sarcoma with limited metastasis at presentation has a poor prognosis.
Our series had 14% of patients (three of 21) with limited metastasis at presentation
who were treated with curative intent. Those who were metastatic at presentation (n = 3) had a survival of 33% at 5 years as against nonmetastatic patients who had 64%
OS at 5 years. Apart from these three, those who developed metastases (seven patients)
subsequently too had poor survival. Of these 10 patients, only two patients are currently
alive (one is disease free 1 year after metastasectomy and the other patient is on
palliative chemotherapy). Our survival of 64% (nonmetastatic patients) is comparable
to other studies by Pradhan et al,[15] who reported a survival of 72% for peripheral lesions and 78% for nonmetastatic
lesions, and Nesbit et al,[16] who reported survival in the range of 56 to 79% (nonpelvis and nonmetastatic).
Poor response to chemotherapy is one of the strongest prognostic factors in Ewing’s
sarcoma.[17] Similarly in our series too, the 5-year OS in 28% (six out of 21) patients with
poor response to chemotherapy was 33% as against the 66% OS in the good responders.
The decision to offer radiotherapy as an adjunct to surgery was decided on a case-by-case
basis as detailed earlier after discussion in a multidisciplinary clinic. Though it
can be argued that there could be an element of subjective clinical bias in the decision
making, consistency was maintained as the treating team was the same for all cases.
There was no statistical significance in local recurrence and survival based on the
addition of radiotherapy. Given that many patients of Ewing’s sarcoma are in the pediatric
age group, our approach in using radiotherapy selectively appears justified. The apparently
better DFS and OS in the group that did not receive radiotherapy is possibly because
of the selection bias; cases that received radiotherapy were judged by the multidisciplinary
team to be in a higher risk group compared with those that did not.
The clavicle being in close vicinity to major anatomical structures can be challenging
to excise especially in cases with a large soft tissue mass. There are differing opinions
regarding the need to reconstruct the clavicle after excision.[1] While several reports have described no abnormality in shoulder function after clavicle
excision, shoulder dysfunction, dyskinesis of the scapula, and damage to the shoulder
joint have been reported.[1]
[18] Though reconstruction using allogeneic and autologous bone, bone cement, and a plate-cement
complex has been done to mitigate post-clavicle excision complications, they in turn
have resulted in increased chances of infection, prosthesis loosening, persistent
pain, and symptomatic nonunion.[1]
[19]
In our series where no reconstruction after excision was attempted a surgical complication
in the form of wound necrosis was observed in only one patient. This too settled with
conservative management without the need for further surgical intervention.
The MSTS score is a well-accepted method to study the functional evaluation of patients
with extremity tumors.[8] Our patients had a mean MSTS score of 29 (range: 27–30). The functional scores were
similar for both the groups (total clavicle excision: 28.5, partial clavicle excision:
29.5). This is in concurrence with the results reported by Kapoor et al,[3] where both total and partial clavicle excision without reconstruction were not associated
with a clinically significant loss of function based on the MSTS score.[3] Wessel and Schaap,[20] in a series of six patients (two malignant tumors) who underwent total clavicle
excision reported that resection of the entire clavicle did not disturb the motion
of the shoulder.[20] In a study by Natarajan et al,[11] the functional outcomes after clavicle excision evaluated in 12 patients (Ewing’s
sarcoma: 6) were excellent in five cases and good in seven cases. The strength of
abduction and flexion though limited in two patients did not impinge on the overall
functional outcomes ([Table 4]).[11]
Our study suffers from the limitations inherent to any retrospective study. The numbers
though small are inevitable given the rarity of the site and the pathology. To the
best of our knowledge this is the largest single institute study of Ewing’s sarcoma
in English literature.
Conclusion
Surgical resection (partial/total clavicle excision) without reconstruction as a part
of multimodality treatment offers acceptable functional and oncological outcomes in
primary Ewing’s sarcoma of the clavicle and the oncologic outcomes are similar to
that of Ewing’s sarcoma at other skeletal sites.
By Dr Govind Babu, President ISMPO and Senior Medical Oncologist
