Oncology

What is the evidence for follow-up in sarcoma?

One of the difficulties with surgical oncology (in any specialty) is, when exactly is follow-up long enough? Patients love to receive the ‘all-clear’ and the funding structures for many modern healthcare systems do not favour specialist follow-up for many conditions – even those that are difficult or impossible to follow up effectively in primary care. To make matters somewhat worse, the majority of follow-up protocols are expert opinion-driven and have not been tested with an evidence base. In an extremely valuable paper, researchers in Toronto, Ontario (Canada) undertook a simple two-part study. Initially they set out to establish what ‘standard of care’ was available in their own local centre based on follow-up frequency for patients, and then go further to establish if this was supported by the evidence for occurrence of new metastases.¹ The authors undertook a careful retrospective review of 680 patients, all presenting at their centre with bone sarcomas. They compared the follow-up protocols in use to the actual outcomes in terms of likelihood of local recurrence and metastasis, and used this to calculate the annual incidence of metastasis. From this, they established that their current follow-up protocols (0-2 years, 2-5 years, 5-10 years and > 10 years) were appropriate, or at least matched with the fall in annual incidence of recurrence and metastasis. This is a really helpful study in so far as it confirms that current follow-up protocols are more or less correct, and offers some guidance on what really should become the standard of care in the future.

Endoprosthesis in the very long term

The endoprosthesis offers a number of quite clear advantages in treating bone tumours: restoration of function, maintenance of joint and limb length and orientation, while facilitating the en bloc excision of bone. Their gains are rapid, and patients return to function in the few weeks following surgery. The worry, of course, is that many bone tumour patients are young, and prosthetic survival in the longer term is complicated by wear, loosening, infection and periprosthetic fracture, all of which become more common year on year. There are few reports of the truly ‘long-term’ follow-up subsequent to the use of endoprosthetic replacement in malignant tumours. In an important study, surgical oncologists from Birmingham (UK) have reported their experiences of the first generation endoprostheses performed for bone tumours using cemented hips and hinged knee arthroplasties.² Their study is one of the few to report patients with more than 25 years of follow-up. The authors were able to identify 230 patients, the majority of whom had osteosarcoma (n = 132), with distal femur being the most common site. Over the nearly 30 years of follow-up reported as part of this study, the authors undertook 610 further operations, the frequency of which was dictated by the presence or absence of infection (4.7 vs 2.1 extra operations). Overall success rates in terms of limb salvage were, however, impressive, with an amputation rate of just 16% at 30 years of follow-up. Nevertheless, with an ongoing year-on-year risk of around 1%, infection rates were not insignificant. The main risk reported in this very long follow-up was infection, which was likely to be partly related to the number of revisions the patients underwent. So it seems that the coast, as they say, is never clear following an endoprosthetic arthroplasty; however, it can be considered a reconstructive choice that for most patients can be a lifelong solution if both patient and surgeon are happy to accept the significant number of re-operations.

Wide local excision means wide local excision

Is delay a problem in bone tumour surgery?

This is another paper highlighting the long delays in diagnosing bone tumours, however, like many other similar papers on the topic, it is not convincingly able to demonstrate that delays correlated with outcome. There is clearly some concern in the medical, scientific and general population about the effects of delay to diagnosis in malignancy. In fact, many countries now see this as a key performance indicator for their health systems. However, the difficulty with a tumour is that it is never clear exactly how long it has been there prior to diagnosis – only how long it has been symptomatic. Researchers from Groningen (The Netherlands) set out to establish what, if any, effect a delay to diagnosis had on outcomes.⁴ Their study revolves around the outcomes of 102 patients, each with a primary bone tumour of differing type (54 osteosarcoma, 29 Ewing’s sarcoma and 19 chondrosarcomas). The diagnosis interval was defined for the purposes of their study as the time between onset of symptoms and establishment of a histological diagnosis, arbitrarily split into short (< four months) and long (> four months) delays. On average, there were longer delays to diagnosis with less aggressive tumours (688 chondrosarcoma vs 160 for osteosarcoma), with most medical delays occurring in the primary care setting. On the face of it, one might expect a delay in diagnosis to have a profound effect. However, overall survival rates were much more determined by oncological diagnosis than by delay to achieving that diagnosis. The explanation for this and other studies that suggest the converse likely lies in the natural history of the disease. Aggressive tumours present earlier as they are more symptomatic – and perhaps not surprisingly, they often fare worse.

Timings of radiotherapy

Sticking with the general theme of timings to treatment in bone tumours, a French multicentre collaborative study based in Lille (France) set out to investigate the association between survival and timings of adjuvant radiotherapy in bone infection surgery.⁵ There is potential to improve outcomes with earlier radiotherapy, however, it profoundly affects wound healing and, particularly where prostheses are involved, one has to be careful that complications associated with early adjacent radiotherapy do not outweigh the potential benefits. These authors set out to establish what, if any, was the impact of adjuvant radiotherapy time interval (TI) between surgery and adjuvant radiotherapy (RT). Their study concerned patients treated with soft-tissue sarcoma over a nearly 15-year period, and reports the outcomes of over 1100 patients. The TI was categorised somewhat arbitrarily (19-39 days, 40-79 days, 80-119 days and > 120 days). The conclusion reached by the authors based on their results was that the timing of radiotherapy does not make a difference to outcome. However, this observation is in direct opposition to reported data for other cancers such as breast, head and neck, where a delay in RT negatively impacts upon outcomes. Certainly food for thought, however, if a delay to radiotherapy is clinically indicated, surgeons and patients should be confident that it will not have too profound a negative impact on outcomes.

Prophylactic fixation: the way forward

The cost and utility of advanced imaging in cartilaginous tumours

The tendency over the past few years has been to move towards increasingly complex imaging modalities as a general rule, especially where there may be diagnostic uncertainty such as in cartilage tumours. This relentless march towards over-investigation is driven by a combination of factors, and it is not always a diagnostic need. Surgeons from Nashville, Tennessee (USA) have set out to establish what, if anything, is the advantage of this approach – are we just needlessly investigating patients?⁸ The authors reviewed the prevalence and cost of advanced imaging studies (AIS) in the evaluation of long bone cartilaginous lesions in 105 enchondromas and 19 chondrosarcomas. They classified advanced imaging as including MRI, CT, bone scan, skeletal survey, or CT biopsy. Of those patients diagnosed with an enchondroma, 85% presented from their primary care physician having already had an advanced imaging study. The average enchondroma patient presented with one unnecessary AIS, with an average unnecessary cost per enchondroma patient of around $1350. The authors make the valid point that, given the 85% concordance between radiologists in their study for enchondromas and 100% concordance for chondrosarcomas, these studies are being requested and performed needlessly. Perhaps most importantly, the authors clearly outline imaging algorithms that could easily and safely be implemented to reduce the unnecessary imaging studies being performed in this setting.

Radiotherapy and limb salvage: a potent combination?

The use of radiotherapy in combination with wide local excision is a well-established treatment for limb salvage and, as the techniques have been refined, has led to survival rates at a similar level as those for radical amputation with a far better quality of life. However, the correct timing and sequence of treatments for radiotherapy and wide local excision remain more than slightly controversial. Researchers in Firenze (Italy) pose, and partially answer, the question: is radiotherapy still needed in patients with high-grade sarcoma and clear wide local margins? Their study describes retrospective outcomes from a mixed cohort of 769 patients, of whom all but 365 were treated with wide local excision and then adjuvant or neo-adjuvant radiotherapy.⁹ The survival rates were slightly different, with the best recurrence-free survivals published at five years of 90% for neo-adjuvant radiotherapy, and 78% for adjuvant. This paper still leaves the question unanswered, primarily because it suffers from the bias that dogs all retrospective studies that span a long period. The enrolment of the patients in the groups was not randomised, and nor was it even throughout the study period. The decision to add radiation therapy was influenced by the size and location of the tumour, as well as the extent of the surgical resection. This, combined with the reported evolution of practice as a shift from adjuvant to neo-adjuvant radiation therapy during the observed study period, leaves us with more questions than answers.