Prognostic significance of tumor size in child osteogenic sarcoma

Significado prognóstico do tamanho tumoral no osteossarcoma infantil

Prognostic significance of tumor size in child osteogenic sarcoma

Significado prognóstico do tamanho tumoral no osteossarcoma infantil

Marcos Almeida Matos^I; Nilma Pimentel^II; Alexandre Alves Leite^III

^IAssistant Professor, Orthopedics at EBMSP

^IIOncologist from Instituto Eric Loeff of HSI

^IIIOrthopedist from Hospital Santa Izabel

Address for correspondence

SUMMARY

From 1995 to 1998, twenty seven pediatric patients were included in a prospective clinical study. Their age ranging from 8.9 to 17.5 years old (mean 13.5), 20 (74%) were male and 7(26%) female, with a follow-up going from 3 to 39 months (average of 17). Osteogenic sarcoma arises in Knee (55.5%), ankle (26%), shoulder (11.1%), pelvis (3.7%) and hip (3.7%). Tumors less than 12cm were considered being small ones, while those bigger than 12cm were considered large. .All patients were submitted to clinical, radiological, histological evaluation and graded by Enneking⁽⁵⁾ system (13 were IIB and 8 were IIIB); treated by chemotherapeutic agents, according to "Brazillian Protocol 91" and had a local surgical control. There were carried out 9 (42.9%) radical surgical interventions and 12 (57.1%) surgeries of limb preservation. According to lesion size, eight (38.1%) out of 14 (67.9%) patients with a tumor greater than 12cm died; in contrast, none out of 7 with tumor less than 12cm relapsed or died. Authors have concluded that tumor size can be a good factor for prognosis indication, being low cost, simple, easy to reproduce and specially useful for taking pre-treatmente decisions.

Key words: Osteosarcoma, prognosis, treatment

RESUMO

Entre 1995 a 1998, vinte e sete pacientes foram incluídos neste estudo clínico prospectivo. A idade variou entre 8,9 e 17,5 anos (média de 13,5), 20 (74%) eram do sexo masculino e 7 (26%) do sexo feminino com seguimento entre 3 e 39 meses (média de 17). O osteossarcoma incidiu no joelho (55,5%), tornozelo (26%), ombro (11,1%0), pelve (3,7%) e quadril (3,7%). Tumores menores que 12cm eram considerados pequenos e maiores que 12cm eram considerados grandes. Os pacientes foram submetidos a avaliação clínica, radiológica e histológica e classificados pelo sistema de Enneking⁽⁵⁾ (13 eram IIB e 8 eram IIIB); foram tratados com agentes quimioterápicos de acordo com o protocolo brasileiro 91 e pelo controle cirúrgico local. Foram realizadas 9 (42,9%) cirurgias radicais e 12 (57,1%) cirurgias de preservação do membro. De acordo com o tamanho tumoral, 8 (38,1%) dos 14 (67,9%) pacientes com tumores maiores que 12cm faleceram; ao contrário, nenhum dos 7 pacientes com tumores menores que 12cm apresentaram recidiva ou faleceram. Os autores concluem que a medida do tamanho tumoral pode vir a ser um bom fator prognóstico, sendo também de baixo custo, simples, de fácil reprodutibilidade e especialmente útil para tomada de decisões pré-tratamento.

Descritores: Sarcoma ósseo, prognóstico, tratamento

INTRODUCTION

The degree of necrosis and tumor size are the only features that in several publications are statistically related to prognosis of patients with osteosarcoma ^{(3,6,7,8,10,11,12,13)}. Degree of necrosis looks to be the most important isolated factor⁽³⁾, however this parameter can only be assessed after local tumor control (surgery), thus, it is not available before surgical decision. Oh the contrary, measurement of tumor size can be performed in a simple manner in pre treatment radiographs, making easier to use this parameter in decision making of the type of surgery to be performed, if radical in worst prognosis patients, or conservative in those with a more optimistic prognosis.

Based on this, it is reported a prospective study aiming to standardize measuring of tumor size in pre treatment radiographs, looking for a relationship of this assessment to short term global survival prognosis free of disease in patients with children osteosarcoma of the extremities.

CASES AND METHODS

A prospective study was performed in twenty seven patients with osteosarcoma of the extremities treated at Hospital Santa Izabel, chosen through sequential sampling method from January 1995 to June 1998 (42 months). All patients underwent complete anamnesis, physical examination, laboratory evaluations (including alkaline phosphatase dosage), radiographic studies, technetium bone scyntilographic evaluation, thoracic CT scan, and CT scan or MRI examination of the affected region. Aiming diagnosis confirmation it was always used an anatomic-pathologic study after a biopsy.

Patient's age ranged from 8.9 to 17.5 years (average 13.5). Twenty (74%) were male and 7 (26%) female (Table 1). Average follow-up was of 17 months, with a minimum of 3 and maximum of 39 months. Knee was affected in 55.5% of the osteosarcomas, and distal femur was involved in 40.8% of the total of lesions (Tables 2 and 3).

Based on hystopathological features, tumors were classified as osteoblastic, teleangectasic and chondroblastic (Table 4). Staging was performed by Enneking's⁽⁵⁾ system.

The size of the tumor was assessed in a routine radiographic exam of the affected bone, in an anterior-posterior view, prior to any treatment. In order to perform the measurement, it was took the length of the tumoral mass in its longitudinal axis, considering the Codmann's triangle as the start of the tumor, and physeal line as its end (Figure 1). In cases where the tumor looks to cross the physeal barrier, it was chosen to use the joint surface for this measurement.

All the patients underwent neo-adjuvant chemotherapy treatment, according to Protocol 91, established by Children Cancer Study Group of Oteogenic Sarcoma⁽²⁾. This protocol consists of 5 cycles of 3 drugs, using ifosfamyde (IFO) at a 1,800 mg/m²/day dose during 5 days each cycle, adryblastin (ADR) at a 25 mg/m²/day dose during 3 days each cycle and cysplatin (CDDP) at a 120 mg/m²/day dose during one day each cycle. All these drugs were administered by continuous intravenous infusion during 4 hours. Cycles were spaced by a 3 weeks interval, and based on the concept of double drug, using IFO-ADR association in the cycles 1^st, 3^rd and 5^th and CDDP-ADR in cycles 2^nd and 4^th.

Three weeks after the last cycle of neo-adjuvant chemotherapy, around the 38^th week of treatment, the patients underwent surgical treatment for local control of the tumor. Choosing of the surgery (radical or conservative) was based on multiple factors, among them the size of the tumor, response to chemotherapy, anatomic location, available technical conditions, stage (patient's prognosis) as well as psychosocial and family factors.

Coadjuvant chemotherapy con-sisted of 7 cycles with a 3 weeks interval, started 3 weeks after surgery. At this moment a 4^th drug, VP-16 was used in two different concentrations. When associated to CDDP, it was used at 150 mg/m²/day during 3 days each cycle, and when associated to IFO, at 100 mg/m²/day during 5 days each cycle. The association VP-CDDP was used in 1^st, 3^rd, 5^th and 7^th cycles, and VP-IFO was used in 2^nd, 4^th and 6^th cycles. In all cycles CDDP and IFO were used at the same dosages and administered according to above described for neo-adjuvant chemotherapy.

Descriptive statistics was used for data presentation and exact Fisher test at a 0.05 level was applied to identify significant differences between treatments.

RESULTS

Of the twenty-seven patients who entered the study, one is still under chemotherapy and five were lost during the follow-up: two moved from the city; two ceased the treatment due to family reasons; one deceased due to complications during coadjuvant chemotherapy. Twenty-one kept available for evaluation of the relationship between size of the tumor and patient prognosis.

After chemotherapic treatment, 57% of the patients had conservative surgeries preserving the limb and 43% underwent amputation or dearticulation (Table 5). Among the conservative surgeries, non conventional endo-prosthesis was used in 67% of the total (Table 6).

Fourteen (67%) patients had lesions larger than 12 cm while 7 (33%) had lesions smaller than 12 cm (Table 8). Thirteen patients (62%) had no metastasis at the time of diagnosis (Enneking IIB) while seven patients already had pulmonary metastasis, and one had bone metastasis (Enneking IIIB). Of the 8 (8%) deaths, one was due to neurological complications, one due to septicemia during chemotherapy course and six due to pulmonary complications (Table 7 and 8).

Eight patients with lesions larger than 12 cm died, while no death was observed among those with tumors smaller than 12 cm. Seven deaths happened in metastatic cases, while only one death was seen among non-metastatic cases (Tables 7 and 8).

DISCUSSION

The presented sample consisted of 27 patients with an average age of 13.5 years, 20 (74%) males and 7 (26%) females. Knee was the most affected location, in 55% of the cases (distal femur in 40.8% and proximal tibia in 14.8%). This profile is in accordance to the evaluated literature, where it is stated that a typical patient with osteosarcoma is among 10 and 20 years old, most frequently male and with a lesion usually located at the knee level ^(1,6,9).

In relation to the surgical treatment that was applied, in (Tables 5 and 6) it can be observed that these data are similar to those presented by the Garcia et al⁽⁶⁾ that reported the occurrence of 62% of conservative surgery versus 37.5% of ablative surgeries (amputations and dearticulations).

Regarding Enneking's⁽⁵⁾ classification, patients were all in groups IIB or IIIB. It can be observed in (Table 8) that global survival was closely related to this classification, that is, an statistically significant difference was found (p = 0.001) between the groups, as the group IIB had a higher global survival (57.1%).

When it is evaluated the relationship among tumor size and patient prognosis, it can be observed that small tumors (less than 12 cm) were 7 cases with no death, while larger tumors (larger than 12 cm) were 14 cases with an incidence of 38.1% deaths (Table 7). It can also be observed an statistically significant difference (p = 0.018) between the groups, leading to believe that tumor size can be a good prognosis factor in evaluation of patients with osteosarcoma. It should also be observed that all 8 deaths occurred among those patients with lesions larger than 12 cm and only in one case it was not a metastatic tumor. This fact indicates a close relationship between size of the tumor and Enneking's⁽⁵⁾ staging, that is, larger tumors (larger than 12 cm) are generally metastatic.

Petrilli⁽¹⁰⁾ stressed that size of the tumor was an important prognosis factor for referencing survival rates of this group of patients. In his paper, he reports that patients with a tumor size of less than 14 cm had a 3 years survival rate of 65%, in contrast to those with tumors larger than 15 cm who had a 3 years survival of 32.3%.

The Garcia et al⁽⁶⁾ also refers to tumor size as a prognosis factor. In this case it was observed that a total remission was found in 76% of the patients with tumors smaller than 12 cm versus an index of 52% among patients with tumors larger than 12 cm. Grades III and IV of necrosis based on Huvos^(8,13) classification, were seen in 63% of the tumors with less than 12 cm (37% were grade I and II) versus only 24% in tumors larger than 12 cm (76% were grade I and II).

Davis⁽³⁾ and cols. reviewed the literature and concluded that only tumor size and tumor necrosis were statistically significant in relation to survival prognosis in more than 50% of the revised papers^{(4,7,10,11,12,14)}. Other variables, according to these authors, had only conflicting influences on prognosis, especially due to existence of a patent lack of standardization of evaluations. The tumor size, however, was a significant prognosis factor in an univariable analysis, however lost its value when the necrosis degree was added in a multivariable analysis. The relationship between the tumor size and degree of necrosis has been studied in experimental models of osteosarcoma and it is possible that large tumors are poor responders to chemotherapy^(3,4,12,14). This could explain why the tumor size becomes of lesser significance as a prognosis factor when tumor necrosis is added to the model in a multivariable analysis. These data also could explain the findings of the Garcia et al⁽⁶⁾ that the tumors with less than 12 cm present a better response to chemotherapy and, in their vast majority, present a more significant necrosis, generally grade II or IV.

The degree of tumor necrosis after chemotherapy described by Huvos and Rosen⁽⁸⁾ looks to be the best prognosis factor related to survival of a patient with osteosarcoma^(8,13,14). This assessment, however, is dependent on histological evaluation of sites retrieved from the tumor piece after surgical treatment for local control. Thus, this method can only be available after surgical treatment. Yet, tumor size can also be a good prognosis factor and is probably related to response to chemotherapy, and consequently to the degree of necrosis. However, this variable presents the advantage of being available since the beginning of the treatment, including before decision making of surgery, for what it could be useful. Another advantage of measurement of tumor size in relation to the degree of necrosis is its simplicity, low cost and reproducibility of this method in plain radiographs.

In this set of cases we opted for using 12 cm as the limit between larger and smaller tumors, based on the works of the Garcia et al⁽⁶⁾ and by Petrilli⁽¹⁰⁾, adding the fact that the smaller tumor size in what a death was observed was 12 cm. So it was observed an strong evidence that the tumor size can be used separately to indicate patient prognosis, particularly in the pre operative period. This parameter did also correlate very well to Enneking's⁽⁵⁾ staging system and the short-term free of disease global survival. It will so, be necessary efforts aiming to perform works with a larger number of cases and with a long-term follow-up for these evidences to be definitely evaluated.

Address for correspondence

Rua Rodolfo Cavalcante, 196, ap. 1701, Jardim Armação

CEP 41750-080 Salvador,BA.

E-mail: [e-mail]

Trabalho recebido em 21/12/2001. Aprovado em 17/05/2002

Work performed at Santa Casa de Misericórdia da Bahia/Hospital Santa Izabel-HSI and Escola Bahiana de Medicina e Saúde Pública-EBMSP - Salvador - BA

• 1. Bacci G., Picci P., Ruggieri P., Mercuri M. et al.: Primary chemotherapy and delayed surgery (neoadjuvant chemotherapy) for osteosarcoma of the extremities. The Instituto Rizzoli experience in 127 patients treated preoperatively with intravenous methotrexate (high versus moderate doses) and intraarterial cisplatin. Cancer 65: 2539-2553,1990.
• ²
  Children Cancer Study Group Osteogenic Sarcoma.: Pilot III protocol. Treatment of newly diagnosed untreated osteosarcoma with ifosfamide, adriamycin and cisplatin as initial therapy. São Paulo, 1992.
• 3. Davis M., Bell R.S., Goodwin P.J.: Prognóstic factors in osteossarcoma: a critical review. J Clin Oncol 12:423431, 1994.
• 4. Eilber F., Giuliano A., Eckardt J., et al.: Adjuvant chemotherapy for osteosarcoma: a randomized prospective trial. J Cli Oncol 5:21-26, 1987.
• 5. Enneking W.F., Spanier S.S., Goodman M.A.: A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop 1980; 153:106.
• 6. Garcia R.J. Consentino E., Camargo O.P. et al.: Tratamento ortopédico do osteossarcoma. Grupo Cooperativo Brasileiro de Tratamento do Osteossarcoma. Rev Bras Ortop 31:871-878, 1996.
• 7. Hudson M., Jaffe M.R., Jaffe N. et al.: Pediatric osteosarcoma: therapeutic strategies, results and prognostic factors derived from a 10-year experience. J Clin Oncol 8: 1988-1997, 1990.
• 8. Huvos A.G., Rosen G., Marcove R.C.: Primary osteogenic sarcoma. Pathologic aspects in 20 patients after treatment with chemotherapy, en bloc ressection, and prosthetic bone replacement. Arch Phatol lab Med 101: 14-18, 1977.
• 9. Matos M.A., Santana F.R., Leite A.A.: Osteossarcoma na infância. Estudo epidemiológico no Estado da Bahia. Rev Bras Ortop 33:739-742, 1998.
• 10. Petrilli A.S., Gentil F.C., Epelman S. et al.: Increased survival, limb preservaion and prognóstic factors for osteossarcoma. Cancer 68:733-737, 1991.
• 11. Petrilli A.S., Penna V., Lopes A., Figueiredo M.T.A., Gentil F.C.: IIB osteosarcoma. Current management, local control, and survival statistics. Clin Orthop 270:60-66, 1991.
• 12. Picci P., Sangiorgi L., Rougraff B.T. et al.: Relationship of chemotherapy-induced necrosis and surgical margins to local recurrence in osteosarcoma. J Clin Oncol 12:2699-2705, 1994.
• 13. Rosen G., Caparros B., Huvos A.G.: Preoperative chemotherapy for osteogenic sarcoma: selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative chemotherapy. Cancer 49: 1221,1982.
• 14. Rosen G., Marcove R.J., Huvos A.G.: Primary osteogenic sarcoma. Eight years of adjuvant chemotherapy. J Cancer Res Clin Oncol 106: 55,1983.

Publication Dates

• Publication in this collection
  25 Feb 2003
• Date of issue
  Sept 2002

History

• Accepted
  17 May 2002
• Received
  21 Dec 2001