1 Introduction

Bone sarcoma remains an extremely rare diagnosis with an incidence of approximately 1 % of all cancers in the adult population. This has been the case since the early 1990s. In the UK there are over 500 new cases per annum with the paediatric and young adult populations account for the majority of cases.¹ This varies according to specific histological subtypes.^2–4 As is the case with many countries in the western hemisphere, the UK has seen an increase in the size of the elderly population. This is reflected in the number of referrals that our centre has received for suspected bone sarcomas in patients aged over 90 years old. While existing guidelines provide protocols for the investigation and management of bone sarcomas for the under 75 population^1,8,9, there is a sparsity of guidance and information pertaining to those patients over 90 years old. Typically this age group is also excluded from medical research and, as such, robust evidence is lacking.^5–7

Survival of 5 years or more in the over 70 population currently stands at 40 % in most recent literature^1,4. This contrasts with a 5 year survival of approximately 70 % in the under 60 population.⁴ Amongst other factors, survival is purported to be influenced by the patient's age at diagnosis, the presence of metastases or local recurrence,¹⁰ and the chosen management strategies. The latter factor also includes the patient's response to chemotherapy.¹² Age at time of presentation is a well-known prognosticator^4,11,13,14. It is, therefore, to be expected that nonagenarians will suffer from the highest 5 year mortality rate. Whether this is due to the disease process or simply advanced age has not been studied.

In contrast to soft tissue sarcoma, chemotherapy and radiotherapy have been shown to be beneficial in the management of certain subtypes of bone sarcoma^15,16 and a good response to non-surgical treatment appears to correlate positively with both reduced local recurrence and enhanced survival rates^9,15,.¹⁶ However, surgical resection remains the mainstay of treatment for most patients. Given the likelihood of multiple co-morbidities and associated frailty in the ageing population, it is a well held hypothesis that elderly patients are treated less aggressively than their younger counterparts^6,17. This is in part due to an underlying misconception that they will be unable to tolerate the same intensive treatment regimens.¹⁸

Evidence contrary to this exists¹⁷ and a recent study looking at the outcomes of soft tissue sarcoma in the over 90's highlights that early and aggressive treatment with appropriate oncology surgery confers the best local recurrence and survival rates in these patients.¹⁹ It is also well established that the elderly populous have poorer outcomes following amputations than the younger population^12,20. This allows limb salvage to be presented as a viable and much more tolerable alternative management in this population. This ultimately allows the patient to have a more dignified quality of life given their inevitably limited life expectancy at this advanced age.²¹

Whilst the evidence in favour of wide margin resections is well established in improving both local recurrence and survival rates,⁴ the consequences of attaining negative margins in the ageing population may prove to be too much in terms of the added morbidity it may bring to the patient post operatively.²⁰ It is, therefore, the opinion of the authors that radical management may not be entirely justified in the presence of other viable alternatives.

In addition, the current guidance⁹ suggests a strict protocol of post-surgical follow up lasting at least a decade. This itself is increased if mega-prostheses or implants are used in limb salvage procedures. Whilst appropriate clinical examination and imaging is invaluable in identifying early signs of recurrent disease, it is difficult to ascertain if this benefit is realised in the nonagenarian population. Given their limited life expectancy, would a less strenuous follow up regime allow this population more dignity in the final years of life?

2 Aim

We present the largest case series of primary bone sarcoma in the over 90's and aim to quantify the local recurrence and survival rates of these patients with a view to rationalising their management and follow-up at our centre.

3 Patients and Methods

Methods: Retrospective interrogation of our prospectively maintained oncology database yielded a list of 15 consecutive patients referred to our centre since 1995 who were diagnosed with bone sarcoma of the axial or appendicular skeleton over the age of 90 years (median 91 years, range 90–100 years). Mean follow-up was 13 months (median 1 month, range 0.5–94 months). Data on demographics (Table 1), management, local recurrence, follow-up, metastasis and survival were collected. Cause of death was identified by review of our database and contacting the patients’ GP surgery.

Surveillance: Post-operative follow-up was carried out in line with internationally agreed guidelines for bone sarcoma. This included clinical examination for local recurrence with subsequent magnetic-resonance imaging based on those examination findings.⁹ Chest radiography was also carried out to identify chest metastatic disease and a subsequent computerised tomography scan performed where indicated. Patients were staged according to the Enneking Staging for Malignant Muscoskeletal Tumours.²² Follow-up occurred at three monthly intervals for the first two years, six monthly up to year five post operatively and annually thereafter.⁹

Statistical Analysis: Statistical analysis was performed using SPSS (Version 24, IBM Corp, Armonk, NY, USA). Distant metastasis-free survival (DMFS) and overall survival (OS) were defined as the time from diagnosis to the date of metastasis detection and date of death from any cause, respectively. Local recurrence-free survival (LRFS) was evaluated as the time from index surgery to the presence of clinically or radiologically detectable local recurrence. Outcomes for living patients were censored at the time of last follow-up. Survival analyses between variables associated with OS and DMFS were performed using the Kaplan Meier method (log rank test). P values < 0.05 were considered statistically significant.

Ethics: The study was conducted with local institutional board approval. Ethical approval was not necessary for the purpose of retrospective service evaluation. All data was kept confidentially and securely in line with the Declaration of Helsinki.

4 Results

4.3

4.3 Survival

At the time of writing, 14/15 patients diagnosed with a bone sarcoma over 90 years of age had died (93 %). Two patients died within 30 days of diagnosis, both of advanced age. One patient is under long term follow-up (see below). The median overall survival was 6 months (95 % CI 0–24 months). The overall survival at 1, 2 and 5 years was 50 % (95 % CI 23–72 months), 29 % (95 % CI 9–52 months) and 21 % (95 % CI 5–45 months) respectively (Fig. 1).

Close

Fig. 1 Kaplan Meier to show overall survival.

Export to PPT

4.3.1

4.3.1 Local recurrence free survival

Median LRFS time was 14 months (range 1–88 months). The LRFS at 2 and 5 years were both 86 % (95 % CI 33–98 months) (Fig. 2). One patient is under long-term follow-up with no detectable local recurrence or metastatic disease following a below elbow amputation for a high-grade chondrosarcoma of the hand.

Close

Fig. 2 Kaplan Meier to show local recurrence free survival.

Export to PPT

4.3.2

4.3.2 Distant metastasis free survival

Median DMFS time was 17 months (range 1–88 months). DMFS at 2 and 5 years were both 85 % (95 % CI 54–96 months Fig. 3).

Close

Fig. 3 Kaplan Meier to show distant metastasis free survival.

Export to PPT

4.3.3

4.3.3 Survival with metastatic disease

One patient had metastatic disease at presentation. One patient developed metastatic disease 3 months after excision with an R0 margin. Median survival time after metastasis was detected was 1.5 months (range 1–2 months). 100 % of patients with metastases were deceased at 1-year. None underwent surgical treatment for their metastatic disease.

4.3.4

4.3.4 Operative vs. non-operative management

The overall survival for patients who underwent surgical intervention was higher than patients who did not undergo surgical intervention. However, it did not reach statistical significance (HR 0.6 [95 % CI 0.22–2) p = 0.51). At 1-year, patients who underwent surgical resection had an overall survival of 63 % (95 % CI 23–86) while patients who were treated non-operatively had an overall survival rate of 33 % (95 % CI 5-68 p = 0.5).

4.3.5

4.3.5 Tumour excision

The overall median survival time for patients who underwent tumour excision with any margin was 13 months (range 2–88 months). Excision carried out with an R0 margin carried a median survival of 22 months (range 2–88 months). Excision with an R2 margin was associated with a survival of 4 months. One patient who underwent tumour excision with an R0 margin developed metastatic disease and died 3 months after surgery.

4.3.6

4.3.6 Amputation

Median survival for patients who underwent amputation was 17 months (range 6–62 months, mean 25 months). None of the patients managed with an amputation developed metastatic disease.

5 Discussion

This study represents the largest case series of primary bone sarcoma in the over 90 population. It has demonstrated that median overall survival of these patients is limited to 6 months (range 0–24 months).

Operative management was associated with longer survival times when compared to non-operative management (median 17 months and 3 months respectively). This corroborates current thinking that early and aggressive treatment affords the best outcomes^8,9. Both operative and non-operative treatments in the context of bone sarcoma are of high intensity and require a level of reserve that the very elderly simply may not have. The benefit of the current multidisciplinary model is that shared decision making allows for fair and objective development of an appropriate management plan considering the patients’ pre-operative mobility, tumour type, stage, location and co-morbidities, amongst other factors. This is reflected in the fact that only 60 % of our patients underwent surgery.

Our data also trended towards improved survival times with amputations versus excision alone (median 17 months and 13 months respectively). When considering limb salvage surgery for bone sarcoma, one must remember that this necessitates wide resection of the affected bone segment. This often leaves significant defects requiring complex reconstruction. Undertaking this level and length of procedure in an elderly, multi-morbid patient is fraught with complications and thus may be considered less appropriate than amputation. One must also consider the lifespan of limb salvage prostheses and the commitment to lifelong follow-up (allowing detection of prosthesis related complications early). This is reflected in our study by the number of patients undergoing amputation instead of tumour excision [Table 1]. However, this can be counter-balanced by the functional impact of surgery. In contrast to soft tissue sarcoma, where a complete muscle compartment may be removed with minimal functional impact, bone sarcoma is more likely to be treated with amputation and thus the functional impact is higher.¹⁸ If a patient is considered fit for surgery but is not overly concerned by length of survival but rather functional impact, and there are indicators that they would fare well with tumour resection, then the authors see no reason why it should not be an option in this patient population, especially given the lower LRR with excision procedures demonstrated in this study. However, if the patient is more concerned with survival time rather than functional impact, then an amputation may prolong survival.

Furthermore, when treated with excision, there was a modest survival advantage afforded by achieving R0 margins (median survival 22 months versus 4 months for R2 margins). Perhaps one of the more surprising findings from our study was the speed at which local recurrences developed despite achieving negative margins. Median LRFS was limited to 14 months (range 1–88 months) potentially due to the collective decision not to pursue chemotherapy in sensitive histiotypes on account of the patient's age and frailty. LRFS was 86 % at 5 years – under current guidelines this would equate to 12 follow-up appointments with subsequent travel implications, radiation exposure from repeated imaging and COVID-19 risk. Despite intensive follow-up, all our patients with local recurrence died within 6 months of having opted to have further surgery.

Similarly, metastatic disease also appeared to develop swiftly despite attainment of negative margins (median DMFS 17 months range 1–88 months). Median survival with metastasis was limited to 1.5 months (range 1–2 months) and carried a 100 % mortality rate at 1-year owing to the fact they had been deemed unfit for further treatment or refused it. 36 % of our patients died due to sarcoma-related disease, 80 % at 1-year.

Overall, this indicates that a pragmatic approach to follow-up is needed. The authors suggest that the time at which the first follow-up is recommended is appropriate at present as it would allow the detection of local recurrence/metastatic disease and treatment if appropriate, however, a judicious approach to continuing follow-up may be considered, especially if no further treatment is planned. GP or local oncology-led follow-up presents an option. Both would allow expedient referral for further treatment without excessive health risk or travel burden to the patient. Alternatively, either would allow prompt referral to local palliative care services should further treatment not be appropriate. Both follow-up strategies would also free up resources at tertiary centres.

6 Conclusion

This study has demonstrated the need for careful consideration of factors other than tumour histology, location and depth in the management of bone sarcoma in a nonagenarian and has highlighted the importance of a patient-centred discussion pre-operatively. Amputation possibly confers a survival advantage versus excision/limb salvage and non-operative treatment but carries a significantly higher functional impact in this population. Rapid development of local recurrence and metastatic disease confirms the appropriateness of the timing of the first follow-up appointment after surgery however, a median overall survival of 6 months suggests that follow-up does not need to be as intensive nor as prolonged in the over 90's population. Therefore, a pragmatic approach to the frequency, length and place of follow-up should be considered. GP and local oncology-led follow-up may be advantageous in this age group.

Guardian/Patient's consent

Guardian/Patient consent was not sought for the purpose of this retrospective service evaluation.

CRediT authorship contribution statement

Laura Jane Hartley: Conceptualization, Methodology, Data curation, Investigation, Formal analysis, Writing – original draft, Writing – review & editing, Project administration. Motaz Al-Aqeel: Methodology, Formal analysis, Writing – review & editing. Naeil Lotfi: Methodology, Writing – review & editing. Martin Goodman: Writing – review & editing, Supervision. Vineet John Kurisunkal: Conceptualization, Methodology, Writing – review & editing, Project administration, Supervision. Scott Evans: Conceptualization, Methodology, Writing – review & editing, Project administration, Supervision.

Institutional Ethical Committee approval

The study was conducted with local institutional board approval. Ethical approval was not necessary for the purpose of retrospective service evaluation. All data was kept confidentially and securely in line with the Declaration of Helsinki.

Funding statement

No funding or sponsorship was received for this research.