Dactinomycin

We studied the bone mineral density (BMD) of 48 long-term survivors of highly malignant osteosarcoma who had been treated according to the chemotherapy protocols of the German- Swiss-Austrian Co-operative Osteosarcoma Study Group which include high-dose methotrexate. The mean age of the patients was 31 6 +/- 4.2 years and the mean follow-up 16 +/- 2.2 years. The BMD of the lumbar spine and of the proximal femur of the non-operated side was measured by dualenergy x-ray absorptiometry. A questionnaire was given to determine life-style factors, medical history and medication. Ten patients were osteoporotic, 21 osteopenic and 17 normal according to the WHO definition.

Eighteen patients suffered fractures after receiving chemotherapy and all had significantly lower levels of BMD for all the sites measured.

J Bone Joint Surg [Br] 2003;85-B:231-7.

Submitted: 13 February 2002; Accepted: I I April 2002

The introduction of preoperative neoadjuvant chemotherapy in the treatment of highly malignant osteosarcoma has improved the outcome.1 The treatment protocols of the German-Austrian-Swiss Co-operative Osteosarcoma Study Group (COSS) have been used for highly malignant osteosarcoma since 1977,1 and it is important to review the longterm side-effects which may result from the disease and its treatment.

Intensive neoadjuvant chemotherapy may affect bone mass.2 Under healthy conditions the bone mineral density (BMD) increases during childhood and adolescence until the peak bone mass is reached.3,4 Any serious disease during the period of bone accumulation, for example, a malignant bone tumour, and its treatment, may predispose patients to methotrexate (MTX) osteopathy, which is characterised by pain in the bones, osteoporosis, and an increase in the risk of fractures Other causes of decreased BMD may be the disease itself, the ectopic production of parathyroid hormone (PTH), paracrine secretion of lymphokines and decreased physical activity.6-8

Studies which have investigated the BMD of long-term survivors of malignant diseases such as leukaemia, have shown normal as well as reduced levels of BMD.9,10 Two have examined the short-term effects of treatment with MTX on the bone mineral content in patients with osteosarcoma and found lower levels in patients who had received high doses of MTX. 11, 12

Our aim was to evaluate the BMD in long-term (>10 years) survivors of highly malignant osteosarcoma which had been treated according to the COSS protocols.

Patients and Methods

Between 1970 and 1990, 262 patients at our institution were diagnosed as having highly malignant osteosarcoma. Of these, 172 received either adjuvant (COSS 77, 1) or neoadjuvant chemotherapy (COSS 80 - COSS 90; 171) according to the COSS protocols which include doxorubicin, highdose MTX, cyclophosphamide, and, additionally, in different combinations bleomycin, dactinomycin, vincristine, cisplatin, and ifosfamide. 1 The study protocol was approved by the Ethics Committee of our institution and all patients gave their informed consent.

From this population, patients were recruited for this study if they had been free from disease for at least ten years after completion of the treatment as shown by radiography and MRI. There were 82 candidates of whom 16 could not be traced. The remaining 66 were contacted by telephone; eight declined to participate. Questionnaires were sent to the remaining 58 patients, to determine demographic and clinical details. Oncological data were obtained from the files of the local Tumour Registry. All 58 patients returned the questionnaires. Functional evaluation was according to Enneking et al,13 histological classification according to Enneking, Spanier and Goodman,14 and histological regression grading (response to preoperative chemotherapy) according to Salzer-Kuntschik, Brand and Delling.15

Of the 58 patients who returned the questionnaire 48 agreed to have bone densitometry performed. These patients form the basis of this study. There were 22 men and 26 women with a mean age of 31.00 +/- 4.24 years (18 to 41). The mean Enneking score for functional evaluation of reconstructive procedures after surgical treatment was 21 + 4.24 points (5 to 29). Tables I and II give details of the patients and ontological data.

All had received chemotherapy and had had resection of the tumour with wide margins. The main site of the tumour was in the femur in 31 patients (four proximal and 27 distal), in the tibia in 13 (11 proximal and two distal), in the proximal humerus in three and in the rib in one. The lesions were solitary in 45 patients and multiple in three. Histological diagnosis revealed highly malignant osteosarcoma in all patients (osteoblastic osteosarcoma in 25, chondroblastic osteosarcoma in eight, fibroblastic osteosarcoma in three, dedifferentiated osteosarcoma in nine, and surface osteosarcoma in three). Five had demonstrable metastases at the time of diagnosis (four in the lung and one in two ribs). In 38 patients Enneking's histological grading was 2b in 31, 2a in five, and la and 3, respectively, in one case each. For 42 patients, the Salzer regression score was grade I to III (good) in 31 (grade I in 18 cases, grade III in 13 cases) and grade IV to VI (bad or no response) in 11 (grade IV in three, grade V in six and grade VI in two).

Neoadjuvant chemotherapy had been given to 47 patients, and adjuvant chemotherapy to one according to the COSS protocol.1 One patient had had additional irradiation.

The limb had been preserved in 36 patients, with an endoprosthetic reconstruction in 31 and en-bloc resection with plating in five. Nine patients had had a Van Ness rotationplasty (eight at the knee, one at the hip). Three had had an amputation, two below the knee and one a disarticulation through the knee. All nine patients with a Van Ness rotationplasty and the three with amputations have external limb prostheses. Of the 31 patients whose limb had been preserved, 28 received a Kotz Modular Femur/Tibia Replacement (KMFR; Howmedica, Kiel, Germany), and three a Howmedica Modular Replacement System (HMRS; Howmedica) in the humerus. There were 21 distal femoral KMFTR prostheses, four in the proximal tibia and three in the proximal femur. Minor or major complications had required treatment in 14 patients.

From the entire group, four patients reported a family history of osteoporosis and three other diseases affecting bones. Sixteen patients had fractures before diagnosis of highly malignant osteosarcoma and 18 patients thereafter. After completion of the chemotherapy, three received or continue to receive treatment with either cortisol or oestrogen replacement therapy. Three had prophylaxis with calcium and five with vitamin D. All reported that they drank milk but at different frequencies; 25 on a daily basis, 16 two or three times a week and seven once a week. Twenty-six patients admitted to regular alcohol intake, 0.5 1 of wine or its equivalent per day/ week for a mean time of ten years (1 to 15), and 24 were smokers or are still smoking and had done so for a mean of ten years (0.8 to 16). Thirty-one took part in a specific physical activity, but at different intensities (2 daily, 8 twice or three times a week, 21 once a week) and 17 did no physical exercise. Three are members of a sports club.

The mean age of the menarche in the 26 women was 13.0 years (9 to 15). In 12 the menstrual cycle had been interrupted by chemotherapy for a mean period of 8.5 months (3 to 18). At the time of follow-up, 23 had regular menstrual cycles and in three, they were irregular. Twenty-one had used oral contraceptives for a mean of five years (0.5 to 14). One has had hormone replacement therapy for a short period. Fourteen and their partners have 16 healthy children. Three women have had elective terminations. Table III gives the details.

Densitometric technique. The BMD of the lumbar spine (LS) and proximal femur (PF) of the non-operated side was measured by dual-energy x-ray absorptiometry (DXA; Hologic QDR 4500; Hologic, Waltham, Massachusetts, Lunar DPXL; Lunar, Madison, Wisconsin). In order to correct for the different techniques we re-calculated the values of the Lunar DXA according to the standardisation approach of Genant et al.16 The raw BMD measurements on the patients were compared with measurements of a reference population and expressed in terms of standard deviations from this reference population (t-scores). A refinement of these standardised measurements results from a comparison of the raw BMD values with those for healthy age- and sex-matched reference subjects. The reference population for age-matched comparisons was based on the National Health and Nutrition Examination Survey (NHANES) database (z-scores).

The patients were classified into three groups according to the t-score of the LS and the PF as recommended by the World Health Organisation guidelines. The overall BMD score was defined as the lower of the LS or PF scores. Osteoporotic patients were defined as those with an overall BMD score of less than -2.5. An overall BMD score of between -2.5 and -1.0 was evidence of osteopenia. An overall BMD score above -1 was regarded as normal.17 The coefficient of variation for short-term precision has been reported to be 0.5% to 1.5% for the BMD of the spinet$ and 1% to 2% for the BMD of the femur.19

Results

Details of the laboratory findings are given in Table IV. There were ten patients with an overall BMD lower than

-2.5 (osteoporotic values), 21 with a BMD between -1.0 and -2.5 (osteopenic values) and 17 with a BMD above -1.0 (normal). The individual scores of the DXA measurements are shown in Table V.

Table VI contains the means, variances and corresponding confidence intervals for the z-scores for each of the four sites as well as the total score. The sample means range from -0.90 to -0.47, and there is strong evidence in at least three instances that the underlying population means are negative (significance level of at least 5%). The sample variances range from 0.92 to 2.14. In four of the five instances there is an increase in variance; in one instance (femoral neck) there is a statistically significant increase (significance level of at least 5%) in the variance suggesting that the magnitude of the impact of therapy on bone density may vary between patients.

Discussion

To our knowledge, our study is the first to determine the BMD in long-term survivors of highly malignant osteosarcoma who received chemotherapy according to the COSS protocol. In these patients the BMD was significantly lower when compared with a young, healthy and age-matched reference population. Based on the WHO guidelines, of the 48 patients, ten showed osteoporotic (20.8%) and 21 osteopenic (43.7%) values. These are unusually high numbers, which is confirmed by the formal analysis. Four of the five bone densitometric sites showed a moderate increase in the standard deviation. In two, intertrochanteric and total femur, the increase was statistically significant and in the femoral neck there was strong evidence of a statistically significant increase.

During childhood and adolescence bone mass is accumulated until the peak bone mass is reached after adolescence.3,4 Peak bone mass is thought to be genetically determined and influenced by a number of endogenous and exogenous factors. Diseases occurring during this critical time period may influence bone accumulation negatively and lead to a lower peak bone mass. A decrease in the BMD of the hip by 1 SD results in a 2.6-fold age-adjusted increase in the risk of future fracture of the hip.21 Therefore, determination of the BMD in long-term survivors of chemotherapy for highly malignant osteosarcoma is obviously important.

Most patients suffering from highly malignant osteosarcoma are young; in our series the age range was between 6 and 20 years. During this period, bone mass accumulates to reach a peak at the late twenties or early thirties. By measuring the BMD in long-term survivors, the impact of (neo)adjuvant chemotherapy in highly malignant osteosarcomas on the accumulation of the bone mass can be studied.

The commonest site of osteosarcoma is the lower limb (more than 90% in our series). It may cause immobilisation and require the use of crutches or a wheelchair for a period of time. Lack of use or immobilisation is known to lead to reduced bone mass. Although the BMD was measured in the lumbar spine and the proximal femur of the contralateral non-operated side, which normally experiences greater loading because of immobilisation of the affected leg, the BMD values of the patients studied were equally distributed over the three WHO groups. Furthermore, there were no differences between patients with limb salvage or Van Ness rotationplasty, and amputation.

Although (neo)adjuvant chemotherapy used in the COSS protocol, focuses on cells of malignant bone tumours, healthy cells are also affected. Of the chemotherapeutic drugs used in the treatment of highly malignant osteosarcoma, MTX is known to be highly cytotoxic to osteosarcoma cells in vivo.22 High concentrations of MTX suppress bone formation by inhibiting the differentiation of early osteoblasts.23 In animal experiments both bone formation and resorption are affected by MTX.24 In patients with leukaemia, treatment with MTX during childhood and adolescence is associated with pain in the bones, osteoporosis and non-traumatic fractures, a syndrome known as MTX osteopathy.5 On the other hand, low-dose MTX, used in the treatment of rheumatoid arthritis, does not seem to affect bone metabolism, either in the short 25,26 or in the long term.27,28

When all available variables for all patients were compared with the results of the DXA measurements, only the following showed a statistically significant correlation: body weight, serum levels of C-telopeptide, types of endoprosthesis and the menarche in female patients. Body weight, but not body mass index, showed a statistically significant correlation with BMD (p = 0.03). Higher values of BMD were seen in patients of greater weight. The effect of weight on BMD seems to be related to the load on weightbearing bones in both sexes. The differences in women after the menopause may be due to the production of oestrogen by adipose tissue.29

Serum levels of bone-specific alkaline phosphatase (an osteoblast enzyme) and osteocalcin (a bone matrix protein) are the best markers of bone formation. The telopeptide markers appear to be the most specific and responsive markers of systemic osteoclast activity.30 In our study the level of C-telopeptides (p = 0.04), but not bone-specific alkaline phosphatase or osteocalcin, showed a significant correlation with the BMD. We therefore speculate that for the patients in this study, lower BMD is related to bone resorption due to greater osteoclast activity, while bone formation seems to be normal (according to serum markers). Although some patients used vitamin D and/or calcium as a prophylactic agent, the serum levels of vitamin D and of calcium were within the normal range.

The type of endoprosthesis was the only variable related to the disease process and its treatment which was significantly associated with the BMD. In the lower limb, proximal femoral and proximal tibial replacements showed lower values of BMD compared with distal femoral replacements and other surgical procedures (Van Ness rotationplasty, enbloc resection and amputation) (p = 0.05). Other variables which we would expect to influence the BMD in patients with highly malignant osteosarcoma, for example, the duration of symptoms before diagnosis or the patients' age at surgery, showed differences between groups of patients with normal BMD and patients with osteopenia or osteoporosis, but these were not significant. Age at surgery, which also means age at chemotherapy, must be relevant because a higher impact on the accumulation of bone mass would be significant the earlier the disturbance occurred. No differences were seen between the different COSS protocols, the level of activity expressed by the functional score according to Enneking et al13 or the histological regression score of Salzer et al. 15

In female patients the age of the menarche correlates significantly with a lower BMD (p = 0.01); the earlier the menarche, the lower is the BMD. Young adult women with a delayed menarche are at risk for irregular menstrual cycles and low peak bone mass.31 Although the absolute numbers of female patients in each group were too small to draw conclusions, there were differences between the mean values for the three WHO groups. Disturbances of the menstrual cycle (irregular cycles at the time of the follow-up examination or interruption during chemotherapy and duration of interruption during chemotherapy) (Table III) are seen to be more common in the group of patients with lower bone mass (osteopenic or osteoporotic). The use of oral contraceptives itself and a longer duration of use seem to have protective effects, which is also seen in young women of reproductive age with a hypo-oestrogenic condition.32 While the number of births is distributed equally across all groups, the number of miscarriages is higher in patients with lower BMD. None of these differences, however, was statistically significant.

We therefore recommend that DXA measurements be performed at the beginning or at least shortly after discontinuation of chemotherapy and, in the case of lower BMD, that follow-up examinations be made every two years thereafter, that calcium and vitamin D prophylaxis be given during or after discontinuation of chemotherapy, and in the case of osteoporotic BMD values, according to the WHO guidelines, that antiosteoporotic treatment be given.

The authors wish to thank F. Auer-Hackenberg, W. Clementi, R. Dieterich, J. Filger-Brillinger, G. Gamillscheg, M. Gartner, E. Geisswinkler, U. Heise, I. Kapfer, G. Kovacs, D. Kornasoff, A. Lang, R. Lehninger, A. Malekpour, W. Mandlik, P. Muhlberghuber, S. Nickl, G. Reuter, H. Pavalec. M. Pohling, G. Reuter, P. Ritschl, H. Rosner, S. Samberger, Ch, Schulze-Bauer, D. Spengler, E. Wellner, and P. Zechner for co-operation in providing clinical data, and S. Blumel, H. Hitschmann. L. Holzer, M. Kevric. N. Pomsar and T. Zettl for help in preparing the manuscript.

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

1. Bielack S, Kempf-Bielack B, Schwenzer D, et al. Neoadjuvant therapy for localised osteosarcoma of extremities: results from the Cooperative Osteosarcoma Study Group. Kin Padiatr 1999;211:260-70.

2. van der Sluis IM, van den Heuvel-Eibrink MM, HAhlen K, Krenning EP, de Muinck Keizer-Schrama SM. Bone mineral density, body composition. and height in long-term survivors of acute lymphoblastic leukemia in childhood. Med Pediatr Oncol 2000;35:415-20.

3. Bonjour JP, Theintz G, Buchs B, Slosman D, Rizzoli R. Critical years and stages of puberty for spinal and femoral bone mass accumulation during adolescence. J Clin Endocrinol Metab 1991;73:555-63.

4. Slosman DO, Rizzoli R, Pichard C, Donath A, Bonjour JP. Longitudinal measurement of regional and whole body bone mass in young healthy adults. Osteoporos It 1994:4:185-90.

5. Stanisavljevic S, Babcock A. Fractures in children treated with methotrexate for leukemia. Clin Orthop 1977; 125:139-44.

6. Ragab AH, Frech RS, Vietti TJ. Osteoporotic fractures secondary to methotrexate therapy of acute leukemia in remission. Cancer 1970;25:580-5.

7. May KP, Mercill D, McDermott MT, West SG. The effect of methotrexate on mouse bone cells in culture. Arthritis Rheum 1996;39:489-94.

8. Zonneveld IM, Bakker WK, Dijkstra PF, et al. Methotrexate osteopathy in long term low-dose methotrexate treatment for psoriasis and rheumatoid arthritis. Arch Dermatol 1996; 132:184-7.

9. Henderson RC, Madsen CD, Davis C, Gold SH. Bone density in survivors of childhood malignancies. J Pediatr Hematol Oncol 1996; 18:367-71.

10. Arikoski P, Komulainen J, Riikonen P, et al. Impaired development of bone mineral density during chemotherapy: a prospective analysis of 46 children newly diagnosed with cancer. J Bone Miner Res 1999; 14:20029.

11. Gnudi S, Butturini L, Ripamonti C, Avella M, Bacci G. The effects of methotrexate (MTX) on bone: a densitometric study conducted on 59 patients with MTX administered at different doses. Ital J Orthop Traumatol 1988;14:227-31.

12. Ripamonti C, Avella M, Gnudi S, Figus E. Effect of high and low doses of methotrexate (MTX) on bone mass in subjects treated for osteosarcoma of the limbs. Minerva Med 1993;84:131-4.

13. Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ. A system for the functional evaluation of reconstructive procedures after surgical treatment of turnouts of the musculoskeletal system. Clin Orthop 1993;286:241-6.

14. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop 1980;153:106-20.

15. Salzer-Kuntschik M, Brand G, Delling G. Bestimmung des morphologischen Regressionsgrades nach Chemotherapie bei malignen Knochentumoren. Pathologe 1983;4:135-41.

16. Genant HK, Grampp S, Glijer CC, et al. Universal standardization for dual X-ray absorptiometry: patients and phantom cross-calibration results. J Bone Miner Res 1994;9:1503-14.

17. Kanis JA, Devogelaer JP, Gennari C. Practical guide for the use of bone mineral measurements in the assessment of treatment of osteoporosis: a position paper of the European foundation for osteoporosis and bone disease. Osteoporos Int 1996;6:256-61.

18. Genant HK, Engelke K, Fuerst T, et al. Noninvasive assessment of bone mineral and structure: state of the art. J Bone Miner Res 1996;11:707-30.

19. Grampp S, Jergas M, Gluer CC, et al. Radiologic diagnosis of osteoporosis: current methods and perspectives. Radiol Clin North Am 1993;31:1133-45.

20. Bickel PJ, Doksum. Mathematical statistics. Vol. 1. 2nd edition. Upper Saddle River, New Jersey: Prentice Hall, 2001.

21. Cummings SR, Black DM, Nevitt MC, et al. Bone density at various sites for prediction of hip fractures. Lancet 1993;341:72-5.

22. Decker S, Winkelmann W, Nies B, van Valen F. Cytotoxic effect of methotrexate and its solvent on osteosarcoma cells in vitro. J Bone Joint Surg [Br] 1999;81-B:545-51.

23. Uehara R, Suzuki Y, Ichikawa Y. Methotrexate (MTX) inhibits osteoblastic differentiation in vitro: possible mechanism of MTX osteopathy. J Rheumatol 2001;28:251-6.

24. Wheeler DL, Vander-Griend RA, Wronski TJ, et al. The short- and long-term effects of methotrexate on the rat skeleton. Bone 1995;16:215-21.

25. Buckley L, Leib E, Cartularo K, Vacek PM, Cooper S. Effects of low-dose methotrexate (MTX) treatment on bone mineral density (BMD) in patients with rheumatoid arthritis. Arthritis Rheum 38 (Suppl):5312.

26. West SG, Hugler R, Battafarano D, McDermott MT, McBiles. Lowdose weekly methotrexate (MTX) does not cause osteoporosis rheumatoid arthritis (RA) patients. Arthritis Rheum 38 (Suppl):5312.

27. Bianchi ML, Cimaz R, Galbiati E, et al. Bone mass change during methotrexate treatment in patients with juvenile rheumatoid arthritis. Osteoporosis Int 1999:10:20-5.

28. Carbone LD, Kaeley G, McKown KM, et al. Effects of long-term administration of methotrexate on bone mineral density in rheumatoid arthritis. Calcif Tissue Int 1999;64: 100-1.

29. Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res 1993;8:567-73.

30. Eyre DR. Bone biomarkers as tools in osteoporosis management. Spine 1997;22(24 Suppl): 17-24.

31. Anai T, Miyazaki F, Tomiyasu T, Matsuo T. Risk of irregular menstrual cycles and low peak bone mass during early adulthood associated with age at menarche. Pediatr Int 2001;43:483-8.

32. Williams JK. Noncontraceptive benefits of oral contraceptive use: an evidence-based approach. Int J Fertil Womens Med 2000;45:241-7.

G. Holzer, P. Krepler, M.A. Koschat, S. Grampp, M. Dominkus, R. Kotz

From Vienna General Hospital, Vienna, Austrio

G. Holzer, MD, Associate Professor

P. Krepler, MD

M. Dominkus, MD, Associate Professor

R. Kotz, MD, Professor and Chairman Department of Orthopaedics

S. Grampp, MD, Associate Professor Department of Radiology, Osteoradiology

University of Vienna, Vienna General Hospital,Wahringer Guirtel 18-20, A1090 Vienna, Austria.

M. A. Koschat, PhD, Associate Professor

New York University, 44 West Fourth Street, New York, NY 10012-1126, USA.

Correspondence should be sent to Dr G. Holzer.

Copyright British Editorial Society of Bone & Joint Surgery Mar 2003
Provided by ProQuest Information and Learning Company. All rights Reserved