1507P - Correlation between tumour necrosis and overall survival in adults with osteosarcoma treated with neoadjuvant cisplatin and doxorubicin

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Anticancer agents
Bone Sarcomas
Biological therapy
Presenter Gareth Rivalland
Authors G. Rivalland, D. Porter
  • Cancer And Blood Service, Auckland City Hospital, 1023 - Auckland/NZ



The degree of tumour necrosis following neoadjuvant chemotherapy has crucial prognostic significance in methotrexate containing regimens. We examined whether a similar relationship exists for a non methotrexate containing regimen in a cohort of adult patients treated at the Auckland Regional Cancer Service between Feb 1996 and Feb 2011.


We retrospectively identified 35 patients with high grade osteosarcoma using data recorded prospectively in the New Zealand Bone Tumour registry and regional pathology laboratories. All patients received neoadjuvant cisplatin and doxorubicin, without high dose methotrexate. Reported necrosis rates following surgery clustered into 3 groups: >90%, 30-65% and <20%. We analysed patients according to poor (<20% necrosis), intermediate (20-90% necrosis) or good (>90% necrosis) prognostic groups. Analysis was on an intention-to-treat basis.


Median overall survival was 130.6 months, with a 5-year survival of 52%. Of the 35 patients, 29% were in the “poor”, 46% in the “intermediate” and 26% in the “good” prognostic groups. Median survival was significantly different between the “good”, “intermediate” and “poor” prognostic groups (“not reached” vs 50.1 mo vs 23.7 mo (logrank test, p = 0.05). 5-year overall survival rate for the “poor”, “intermediate” and “good” groups was 23%, 49% and 83% respectively. 16 patients (46%) relapsed; 13 patients relapsed (81%) in the lungs alone, 1 local recurrence, 2 lung and other sites. 8 underwent metastasectomy with a median survival of 50 mo vs 21 mo in those whose disease was not operable (p = 0.0009).


This study demonstrates a clear trend to better outcomes with greater degrees of tumour necrosis, but challenge the dogma that necrosis of less than 90% confers a uniformly poor prognosis. Those in the “intermediate” and “poor” response groups had lower 10-year survival rates, but there were long-term survivors in both groups. Overall survival in the “poor” response group is similar to that of historical controls treated by surgery alone, despite a change of chemotherapy in 40% of this group Those with relapsed disease have significant benefit from surgery for metastases.


D. Porter: Previous consulting fro Pfizer. Current consulting for GSK. Previous travel grant from Novartis.

All other authors have declared no conflicts of interest.