u
EDITORIAL
Bone Involvement in Multiple Myeloma ROBERTS.WEINSTEIN,M.D., Augusta,
Georgia
T
he combination of aching bones and pathologic fractures, Bence Jones proteinuria, and progressive cachexia has indicated multiple myeloma since Kahler first emphasized the diagnostic criteria in 1669. Snapper [l] later recognized that pathologic fracture of the sternum is practically pathognomonic of multiple myeloma. Although chemotherapy for multiple myeloma has prolonged the survival period [2], progressive osteolytic lesions, spontaneous fractures, and incapacitating bone pain remain major problems in the clinical management of the disease. Bone pain is the most frequent complaint in patients with multiple myeloma [3] and is the presenting feature in 63% to 90% [1,3,4]. The pain, which nearly always starts in the lower back and ribs, is aggravated by the muscular effort of a mere cough or a sneeze. In the later stages of the disease, severe pain may occur with the slightest physical motion. The report by Mariette et al [5] in this issue of The American Journal of Medicine addresses the impact of chemotherapy on the bone involvement in multiple myeloma. Before this report can be carefully examined, the pathogenesis, evaluation, and treatment of the bone disease should be briefly reviewed. The bone disease develops for several reasons. The proliferating B-cell monoclone can synthesize and release soluble factors that stimulate osteoclasts to resorb bone [6]. These osteoclast-activating factors are primarily composed of interleukinl/3 with a smaller contribution from tumor necrosis factor-8 (previously known as lymphotoxin) [7-lo]. The direct relationship between the extent of skeletal destruction, the tumor-cell burden, and the amount of osteoclast-activating factors produced by cultured myeloma cells indicates that bone involvement in multiple myeloma is due to the production of these bone-resorbing cytokines [ll]. Bone biopsies from patients with multiple myeloma show increased numbers of osteoclasts in the areas invaded by plasma cells [12]. Further aggravation of the bone disease occurs because the cytokines produced by myeloma cells From the Metabolic Bone Disease Laboratory, Section of Metabolic and Endocrine Disease, Department of Medicine, Medical College of Georgia, Augusta, Georgia. Requests for reprints should be addressed to Robert S. Weinstein, M.D., Medical College of Georgia, HB-5027, Augusta, Georgia 30912. Manuscript submitted July 29,1992, and accepted August 19. 1992.
December
also inhibit bone formation [lo]. Therefore, the osteolytic lesions lack reactive or compensatory osteoblast proliferation. For this reason, the serum alkaline phosphatase activity is usually normal unless liver damage is present [l]. Advanced osteolytic disease is associated with elevated urinary excretion of hydroxyproline [13] and low levels of serum osteocalcin [14], representing the increase in osteoclasts and decrease in osteoblasts [9,10]. Hypercalcemia develops in at least a third of patients with myeloma and most probably occurs in those with a larger tumor-cell burden and greater release of cytokines [6]. The hypercalcemia is accompanied by subnormal levels of intact parathyroid hormone (PTH), as measured by the new twosite immunoradiometric methodology, which detects only the bioactive, secretory form of the hormone [15]. Low levels of serum 1,25-dihydroxyvitamin Ds have also been reported [16]. The osteolytic lesions along with the low PTH secretion usually result in high-normal or elevated levels of serum inorganic phosphorus [l]. Renal impairment, which may eventually occur in about 50% of patients, is also associated with hyperphosphatemia. If a carboxyl terminal or middle molecule-specific PTH assay is used in hypercalcemic patients with renal failure, a modestly elevated PTH value may be found. These assays are notoriously unreliable in renal insufficiency because elevated levels are almost always detected and represent decreased renal clearance of PTH fragments rather than increased PTH secretion [17]. When hypercalcemia is found in myelomatous patients, associated bone pain, weight loss, and anemia are expected [3]. Ionized hypercalcemia may be overlooked in patients with advanced disease who have unremarkable serum total calcium concentrations due to concurrent hypoalbuminemia [18] or misdiagnosed in those rare patients with IgG-K myeloma proteins that abnormally bind calcium and cause misleadingly high serum total calcium levels [19,20]. Therapy of multiple myeloma with prednisone may affect the bone disease. Glucocorticoids inhibit intestinal calcium absorption and increase urinary calcium excretion, resulting in a negative calcium balance [21]. In patients with multiple myeloma, glucocorticoid therapy further decreases osteoblast vigor and bone formation. Bone resorption is enhanced by glucocorticoids, but this is primarily due 1992
The American
Journal
of Medicine
Volume
93
591
BONE INVOLVEMENT IN MULTIPLE MYELOMA / WEINSTEIN
to the secondary hyperparathyroidism caused by the negative calcium balance. Nevertheless, it seems unlikely that PTH contributes to bone resorption in patients with multiple myeloma. Glucocorticoids may also paradoxically inhibit bone resorption that has been stimulated by interleukin and tumor necrosis factor [21]. The combined impact of these glucocorticoid effects on the bone mass in multiple myeloma is, therefore, hard to predict. Another possible factor in the bone disease is the decrease in gonadal function that occurs with glucocorticoid treatment [22], chemotherapy [23], or severe systemic disease [24]. Reductions in the secretion of sex hormones may increase the rate of bone loss. Gonadotropin and sex hormone levels should be measured in patients with multiple myeloma because of the potential benefit of gonadal hormone replacement therapy on the skeleton. In addition to these factors, the decreased ambulation of the patient with severe pain may also contribute to the loss of bone mass [21]. Roentgenographic examination of the skeleton yields abnormal findings in approximately 80% of patients [4]. In more than half, there is a combination of lytic lesions, pathologic fractures, and a generalized increase in skeletal radiolucency [3,4]. Multiple myeloma may, however, present with osteopenia alone in 10% of patients and thus masquerade as postmenopausal osteoporosis or senile osteoporosis, if monoclonal gammopathy and marrow plasmacytosis are not discovered [3,4]. Skeletal surveys remain negative in up to 20% of patients [3,4]. The increased sensitivity of computed tomography (CT) may identify focal lytic lesions of the spine in patients with back pain, increased plasma cells in the bone marrow, and a monoclonal protein in the serum but no radiologic changes of myeloma [25]. Magnetic resonance imaging may prove to be even more valuable for the early diagnosis of myeloma bone disease [26]. When lytic lesions are identified, they are most commonly found in the skull, ribs, clavicle, sternum, spine, pelvis, and proximal parts of the extremities [1,3]. The sharply demarcated, punched-out, sieve-like appearance of the calvarium is particularly well known. However, it should be realized that similar radiographic abnormalities may be caused by metastatic carcinoma, especially of the breast [1,3]. With vertebral collapse due to multiple myeloma, the intervertebral disk usually remains intact [l] and the vertebral pedicles are rarely involved, reflecting the relative paucity of red marrow in the pedicles as compared with the vertebral body [27]. Preservation of the pedicles, an intact intervertebral disk, and the more frequent presence of a paraspinal mass are guides that help distinguish multiple myeloma from metastatic car592
December 1992 The American Journal of Medicine
cinoma [1,3]. Because of the defective osteoblast response to the osteolysis, the lytic lesions receive poor uptake of radionuclides [28]. For this reason, bone scans generally underestimate the number of lytic lesions in patients with multiple myeloma. Osteosclerotic lesions occur in only about 3% of patients but can result in positive bone scans [29]. Osteosclerotic myeloma may be associated with chronic progressive polyneuropathy, hepatosplenomegaly, diabetes mellitus, and gynecomastia [30]. A single, particularly painful bone lesion can be palliated with moderate doses of local radiotherapy. Radiotherapy must, however, be limited to a small field and adjunctive chemotherapy delayed until the hematologic impact of the radiation is clear. Lesions usually persist, and radiographic evidence of skeletal repair is rare. Regression of lytic lesions on serial lateral skull roentgenograms has, however, been demonstrated in 30% of patients who respond to melphalan with a reduction in myeloma protein production [31]. Consequently, many attempts have been made to develop adjunctive therapy to treat or prevent widespread bone disease in multiple myeloma. In a large double-blind cooperative study, fluoride therapy was shown to be ineffective and possibly detrimental [32]. Another study examined a regimen of calcium carbonate combined with sodium fluoride, but there were no significant increases in bone density as measured by single-photon absorptiometry of the radial diaphysis, a cortical bone site, or of the distal radius, a primarily cancellous bone site [33]. Dichloromethylene bisphosphonate, a bone-seeking inhibitor of osteoclasts, decreases urinary calcium and hydroxyproline excretion in patients with active myeloma and, in some patients, has diminished bone pain [34,35]. This potentially useful drug is, however, not available in the United States. Even more promising are the reports of inhibition of osteolytic bone lesions with the more potent and longer-lasting agent, (3amino-1-hydroxypropylidene)1, l-bisphosphonate or APD, currently available as pamidronate disodium (Aredia, CIBA-GEIGY, Summit, NJ) [36,37]. In one report, the relief of bone pain with pamidronate lasted 10 to 24 months, but evidence of sclerosis within a previously lytic lesion occurred only once [38]. The side effects of this drug are confined to a transient increase in temperature of 1°C to 2OC and infrequent, mild, local irritation at the injection site. In short-term studies, salmon calcitonin has also been shown to decrease bone resorption and hypercalcemia in patients with multiple myeloma [39]. The hypocalcemic response to calcitonin may even be used to estimate the extent of lytic lesions [40]. Further studies are in progress to assess the value of pamidronate and calcitonin in
Volume 93
BONE INVOLVEMENT
the prevention and treatment of osteolytic lesions in multiple myeloma. Considerable interest has been focused on recalcification of lytic lesions as proof of a beneficial drug response, although this is not a particularly sensitive assessment of bone healing. Estimates of the amount of bone mineral that must be lost or gained to be evident to the trained radiologist range from 20% to 60%. Mariette and associates [5] analyzed bone density determinations in 10 patients with stage III myeloma. Stage III myeloma is characterized by a high tumor burden (greater than 1.2 X 1012 myeloma cells/m2) with a hemoglobin level less than 8.5 g/dL, a serum calcium level greater than 12 mg/dL, advanced lytic bone disease, or high M-component production rates [41]. The patients were treated with high-dose melphalan, carmustine, cyclophosphamide, and etoposide, total body irradiation, and autologous blood stem cell transplantation. Between stem cell collection by leukopheresis and autograft, the patients received three monthly courses of vincristine, Adriamycin, and methylprednisolone. Bone mineral density was measured at the lumbar spine and femoral neck by dual-energy radiograph absorptiometry (DXA or DEXA) at the time of diagnosis of the myeloma and 8 to 12 months after therapy. DEXA uses a stabilized radiographic tube to generate photons of two different energies (one photon is attenuated by bone, the other by soft tissue) and allows a much greater photon flux than with decay of nuclear sources. This state-of-the-art technique thus increases the image resolution and precision while minimizing the radiation exposure and time required for the measurements. Excluding one patient with profound initial osteopenia and an extraordinary gain in bone density after therapy, the remaining nine patients had a mean increase in lumbar density of 0.057 g/cm2, representing a gain of 7.7% (95% confidence interval, 2.2% to 12.2%). At the femoral neck, the mean difference was only -0.018 g/cm2 or -2.3%. Although the mean change at the femoral neck was not statistically significant, the 95% confidence interval was -7.3% to 2.8%, indicating that some patients lost a considerable amount of bone from the hip. Routine radiographic, CT, and magnetic resonance imaging examinations were done at the same intervals, and results did not noticeably change. With the DEXA procedure, Mariette and coworkers [5] found that the long-term coefficient of variation with a phantom of the lumbar spine was 0.5%. Short-term variability in four young volunteers with normal bone density was even better (0.4%) [42]. Even if these precision studies were done by the same technicians with the same densitometer, additional information on the clinical reliDecember
IN MULTIPLE
MYELOMA
/ WEINSTEIN
ability of the method is desirable. The patients with multiple myeloma were, after all, older and more osteopenic. In addition, the interval between the repeat measurements in the patients was about 1 year. A fairer comparison would involve the longterm reproducibility of the densitometric determinations in subjects of similar age and bone density as the experimental group. In the Metabolic Bone Disease Laboratory of the Medical College of Georgia, 56 white women, aged 45 to 84 years with a mean lumbar mineral density of 0.739 g/cm2, had repeat DEXA measurements made over a mean interval of 23 days (range: 2 to 68 days). The 95% confidence limit for the difference in the measurements of Ll-L4 was 0.032 g/cm2. Application of this rigorous test to the results of Mariette and associates [5] indicates that the increases in lumbar density found in the patients with multiple myeloma could be dependably detected by the DEXA technique. Finally, as to whether the increased density was due to the treatment regimen or the DEXA measurements, an instructive patient was recently seen at the Medical College of Georgia. A 53-year-old white man with multiple myeloma received conventional melphalan and prednisone therapy over a 9month interval. DEXA measurements of the lumbar spine increased from 0.724 to 0.790 g/cm2, representing a gain of 0.074 g/cm2 or 9.1%, whereas at the femoral neck, density decreased from 0.622 to 0.533 g/cm2, indicating a loss of 0.089 g/cm2 or 14.3% (95% confidence limit at the hip, 0.039 g/cm2). These findings are similar to those reported by Mariette et al [5] and suggest that the improvement in lumbar density is detectable because of the sensitivity of the new DEXA technique, rather than from more aggressive therapy of the myeloma. Furthermore, the simultaneous decrease in bone density at the femoral neck implies an increasing risk of hip fracture and stresses the vital importance of performing bone densitometry at multiple skeletal sites. The axial and appendicular bones respond differently to various systemic conditions and drug treatments. Augmentation of bone density in response to therapy is usually more conspicuous in the spine than in the appendicular skeleton [43]. The last sentence in the paper by Mariette and associates [5] is possibly the most important. Bone densitometry could well become a routinely measured prognostic factor in the treatment of patients with multiple myeloma. Now that the bone involvement can be precisely assessed with DEXA and promising therapeutic options with calcitonin or pamidronate are available [36-391, prevention of skeletal pain and fracture in multiple myeloma should no longer be considered impossible. 1992
The American
Journal
of Medicine
Volume
93
593
BONE INVOLVEMENT
IN MULTIPLE
MYELOMA
/ WEINSTEIN
REFERENCES 1. Snapper I. Bone diseases in medical practice.
New York: Grune & Stratton,
1957: 157-68. 2. Gregory WM. Richards MA. Malpas JS. Combination chemotherapy versus melphalan and prednisone in the treatment of multiple myeloma: an overview of published trials. J Clin Oncol 1992; 10: 334-42. 3. Kyle RA. Multiple myeloma: review of 869 cases. Mayo Clin Proc 1975; 50:
23. Warne GL. Fairley KF, Hobbs JB. Martin FIR. Cyclophosphamide-induced ovarian failure. N Engi J Med 1973; 289: 1159-62. 24. Gebharl SSP. Watts NB, Clark RV. Umpierrez G, Sgoutas D. Reversible impairment of gonadotropin secretion in critical illness. Arch Intern Med 1989;
149: 1637-41. 25. Soloman A, Rahamani R. Seligsohn U, Ben-Artzi involvement assessed by computerised tomography.
F. Multiple myeloma: early Skeletal Radio1 1984; 11:
29-40.
254-61.
4. Kapadia SB. Multiple myeloma: a clinicopathologic study of 62 consecutively autopsied cases. Medicine (Baltimore) 1980; 59: 380-92. 6. Mariette X, Khalifa P. Ravaud P, et a/. Bone densitometry in patients with multiple myeioma. Am J Med 1992; 93: 595-8. 6. Mundy GR, lbbotson KJ. D’Souza SM. Simpson EL, Jacobs JW. Martin TJ. The hypercalcemia of cancer. N Engl J Med 1984; 310: 1718-28. 7. Stashenko P. Dewhirst FE, Peros WJ. Kent RL, Ago JA. Synergistic interactions between interleukin 1. tumor necrosis factor, and lymphotoxin in bone resorption. J lmmunol 1987; 138: 1464-8.
26. Ludwig L. Tscholakoff D, Neuhold A, Fruhwald F. Rasoul S, Fritz E. Magnetic resonance imaging of the spine in multiple myeloma. Lancet 1987; 2: 364-6. 27. Jacobson HG. Poppel MH, Shapiro JH. Grossberger S. The vertebral pedicle sign: a roentgen finding to differentiate metastatic carcinoma from multiple myeloma. AIR Am J Roentgen01 1958; 801 817-21. 28. Wahner HW. Kyle RA. Beabout JW. Scintigraphic evaluation of the skeleton in multiple myeloma. Mayo Clin Proc 1980; 55: 739-46. 29. Frame B, Honasoge M. Kottamasu SR. Osteosclerosis. hyperostosis and related disorders. New York: Elsevier, 1987: 7&3. 30. lmawari M. Akatsuka N. lshibashi M. Beppu H, Suzuki H, Yoshitoshi Y. Syndrome of plasma cell dyscrasia, polyneuropathy and endocrine disturbance. Ann Intern Med 1974; 81: 490-3. 31. Rodriguez LH. Finkelstein JB, Shullenberger CC, Alexanian R. Bone healing in multiple myeloma with melphalan chemotherapy. Ann Intern Med 1972; 76:
6. Garrett IR, Durie BGM. Nedwin GE, eta/. Production of lymphotoxin, a boneresorbingcytokine, by cultured human myeloma cells. N Engl J Med 1987; 317:
526-32. 9. Bertolini DR. Nedwin GE, Bringman TS, Smith DD, Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 1986; 319: 516-8. 10. Stashenko P. Dewhirst FE, Rooney ML, Desjardins LA, Heeley JD. Interleukin-la is a potent inhibitor of bone formation in vitro. J Bone Miner Res 1987; 2:
559-65. 11. Durie BGM, Salmon SE, Mundy GR. Relation of osteoclast production
to extent of bone disease in multiple myeloma.
activating factor Br J Haematol 1981;
47: 21-30. 12. ValentinQpran tive histology
A, Charhon SA. Meunier PJ, Edouard CM. Arlot ME. Quantitaof myelomainduced bone changes. Br J Haematol 1982; 52:
601-10. 13. Niell HB. Neely CL, Palmieri GM. The postabsorptive
551-6. 32. Harley JB. Schilling A, Glidewell 0. Ineffectiveness of fluoride therapy in multiple myeloma. N Engl J Med 1975; 286: 1283-8. 33. Kyle RA. Jowsey J, Kelly PJ. Taves DR. Multiple myeloma bone disease. N Engl J Med 1975; 293: 1334-8. 34. Siris ES, Sherman WH, Baquiran DC, Schlatterer JP, Osserman EF, Canfield RE. Effects of dichloromethylene diphosphonate on skeletal mobilization of calcium in multiple myeloma. N Engl J Med 1980; 302: 310-5. 35. Delmas PD. Charhon S, Chapuy MC, et a/. Long-term effects of dichloromethylene diphosphonate (ClzMDP) on skeletal lesions in multiple myeloma. Metab Bone Dis Rel Res 1982; 4: 163-8. 36. van Breukelen FJM, Bijvoet OLM, van Oosterom AT. Inhibition of osteolytic bone lesions by (5amino-1-hydroxypropylidene)-1, I-bisphosphonate (A.P.D.). Lancet 1979; 1: 803-5. 37. Sleeboom HP, van Oosterom AT, Bijvoet OLM, Gleed JH, O’Riordan JLH. Comparison of intravenous (3-amino-1-hydroxypropylidene)-1. l-bisphosphonate and volume repletion in tumor-induced hypercalcemia. Lancet 1983; 2:
urinary hydroxyproline (spot-hypro) in patients with multiple myeloma. Cancer 1981; 48: 783-7. 14. Carlson K, Ljunghall S. Simonsson B, Smedmyr B. Serum osteocalcin concentrations in patients with multiple myeloma-correlation with disease stage and survival. J Intern Med 1992; 231: 133-7. 15. Nussbaum SR. Zahradnik RJ, Lavigne JR, et a/. Highly sensitive two-site immunoradiometric assay of parathyrin. and its clinical utility in evaluating patients with hypercalcemia. Clin Chem 1987: 33: 1364-7. 16. Lawson-Matthew PJ. Clayton JC, Guilland-Cumming DF, et a/. Defective production of 1,25(OH)zDs in myeloma. In: Norman AW, Schaefer K. Grigoleit HG, Herrath Dv. editors. Vitamin D: a chemical, biochemical and clinical update. Berlin: Walter de Gruyter & Co, 1985: 891-2. 17. Measuring the PTH level [editorial]. Lancet 1988; 1: 94-5. 18. Chen Y-H, Magalhaes MC. Hypoalbuminemia in patients with multiple myeloma. Arch Intern Med 1990; 150: 605-10. 19. Duggan DB. Schattner A. Unusual manifestations of monoclonal gammopathies: autoimmune and idiopathic syndromes. Am J Med 1986; 81: 864-70. 26. Nora NA. Singer I. Interpretation of hyparcalcemia in a patient with endstage renal disease. Arch Intern Med 1992; 152: 1321-2. 21. Lukert BP, Raisz LG. Glucocorticoid-induced osteoporosis: pathogenesis and management. Ann Intern Med 1990; 112: 352-64. 22. Ma&dams MR, White RH, Chipps BE. Reduction of serum testosterone levels during chronic $ucocorticoid therapy. Ann Intern Med 1986; 104:
du contenu mineral osseux par radiographie digitale quantitative. Premiers resultats au niveau vertebral lombrire. Presse Med 1989; 18: 1062-5. 43. Need AC, Nordin BEC. Which bone to measure? Osteoporosis Int 1990; 1:
648-51.
3-6.
594
December
1992
The American
Journal
of Medicine
Volume
239-43. 36. Thiebaud D. Leyvraz S, von Fliedner V, et a/. Treatment of bone metastases from breast cancer and myeloma with pamidronate. Eur J Cancer 1991; 27:
3741. 36. Rico H, Hernandez ER. Diaz-Mediavilla J. Alvarez A, Martinez R. Espinos D. Treatment of multiple myeloma with nasal spray calcitonin: a histomorphometric and biochemical study. Bone Miner 1990; 8: 231-7. 46. Bataille R, Sany J. Clinical evaluation of myeloma osteoclastic bone lesions: II. Induced hypocalcemia test using salmon calcitonin. Metab Bone Dis Rel Res
1982; 4: 39-42. 41. Salmon SL. Multiple myeloma. In: Wyngaarden JB, Smith LH, editors. Cecil textbook of medicine. 16th ed. Philadelphia: WB Saunders, 1982: 965-70. 42. Braillon P, Duboeuf F, Meaty MF, Barret P, Delmas PD. Meunier PJ. Mesure
93
EDITORIAL
Bone Involvement in Multiple Myeloma ROBERTS.WEINSTEIN,M.D., Augusta,
Georgia
T
he combination of aching bones and pathologic fractures, Bence Jones proteinuria, and progressive cachexia has indicated multiple myeloma since Kahler first emphasized the diagnostic criteria in 1669. Snapper [l] later recognized that pathologic fracture of the sternum is practically pathognomonic of multiple myeloma. Although chemotherapy for multiple myeloma has prolonged the survival period [2], progressive osteolytic lesions, spontaneous fractures, and incapacitating bone pain remain major problems in the clinical management of the disease. Bone pain is the most frequent complaint in patients with multiple myeloma [3] and is the presenting feature in 63% to 90% [1,3,4]. The pain, which nearly always starts in the lower back and ribs, is aggravated by the muscular effort of a mere cough or a sneeze. In the later stages of the disease, severe pain may occur with the slightest physical motion. The report by Mariette et al [5] in this issue of The American Journal of Medicine addresses the impact of chemotherapy on the bone involvement in multiple myeloma. Before this report can be carefully examined, the pathogenesis, evaluation, and treatment of the bone disease should be briefly reviewed. The bone disease develops for several reasons. The proliferating B-cell monoclone can synthesize and release soluble factors that stimulate osteoclasts to resorb bone [6]. These osteoclast-activating factors are primarily composed of interleukinl/3 with a smaller contribution from tumor necrosis factor-8 (previously known as lymphotoxin) [7-lo]. The direct relationship between the extent of skeletal destruction, the tumor-cell burden, and the amount of osteoclast-activating factors produced by cultured myeloma cells indicates that bone involvement in multiple myeloma is due to the production of these bone-resorbing cytokines [ll]. Bone biopsies from patients with multiple myeloma show increased numbers of osteoclasts in the areas invaded by plasma cells [12]. Further aggravation of the bone disease occurs because the cytokines produced by myeloma cells From the Metabolic Bone Disease Laboratory, Section of Metabolic and Endocrine Disease, Department of Medicine, Medical College of Georgia, Augusta, Georgia. Requests for reprints should be addressed to Robert S. Weinstein, M.D., Medical College of Georgia, HB-5027, Augusta, Georgia 30912. Manuscript submitted July 29,1992, and accepted August 19. 1992.
December
also inhibit bone formation [lo]. Therefore, the osteolytic lesions lack reactive or compensatory osteoblast proliferation. For this reason, the serum alkaline phosphatase activity is usually normal unless liver damage is present [l]. Advanced osteolytic disease is associated with elevated urinary excretion of hydroxyproline [13] and low levels of serum osteocalcin [14], representing the increase in osteoclasts and decrease in osteoblasts [9,10]. Hypercalcemia develops in at least a third of patients with myeloma and most probably occurs in those with a larger tumor-cell burden and greater release of cytokines [6]. The hypercalcemia is accompanied by subnormal levels of intact parathyroid hormone (PTH), as measured by the new twosite immunoradiometric methodology, which detects only the bioactive, secretory form of the hormone [15]. Low levels of serum 1,25-dihydroxyvitamin Ds have also been reported [16]. The osteolytic lesions along with the low PTH secretion usually result in high-normal or elevated levels of serum inorganic phosphorus [l]. Renal impairment, which may eventually occur in about 50% of patients, is also associated with hyperphosphatemia. If a carboxyl terminal or middle molecule-specific PTH assay is used in hypercalcemic patients with renal failure, a modestly elevated PTH value may be found. These assays are notoriously unreliable in renal insufficiency because elevated levels are almost always detected and represent decreased renal clearance of PTH fragments rather than increased PTH secretion [17]. When hypercalcemia is found in myelomatous patients, associated bone pain, weight loss, and anemia are expected [3]. Ionized hypercalcemia may be overlooked in patients with advanced disease who have unremarkable serum total calcium concentrations due to concurrent hypoalbuminemia [18] or misdiagnosed in those rare patients with IgG-K myeloma proteins that abnormally bind calcium and cause misleadingly high serum total calcium levels [19,20]. Therapy of multiple myeloma with prednisone may affect the bone disease. Glucocorticoids inhibit intestinal calcium absorption and increase urinary calcium excretion, resulting in a negative calcium balance [21]. In patients with multiple myeloma, glucocorticoid therapy further decreases osteoblast vigor and bone formation. Bone resorption is enhanced by glucocorticoids, but this is primarily due 1992
The American
Journal
of Medicine
Volume
93
591
BONE INVOLVEMENT IN MULTIPLE MYELOMA / WEINSTEIN
to the secondary hyperparathyroidism caused by the negative calcium balance. Nevertheless, it seems unlikely that PTH contributes to bone resorption in patients with multiple myeloma. Glucocorticoids may also paradoxically inhibit bone resorption that has been stimulated by interleukin and tumor necrosis factor [21]. The combined impact of these glucocorticoid effects on the bone mass in multiple myeloma is, therefore, hard to predict. Another possible factor in the bone disease is the decrease in gonadal function that occurs with glucocorticoid treatment [22], chemotherapy [23], or severe systemic disease [24]. Reductions in the secretion of sex hormones may increase the rate of bone loss. Gonadotropin and sex hormone levels should be measured in patients with multiple myeloma because of the potential benefit of gonadal hormone replacement therapy on the skeleton. In addition to these factors, the decreased ambulation of the patient with severe pain may also contribute to the loss of bone mass [21]. Roentgenographic examination of the skeleton yields abnormal findings in approximately 80% of patients [4]. In more than half, there is a combination of lytic lesions, pathologic fractures, and a generalized increase in skeletal radiolucency [3,4]. Multiple myeloma may, however, present with osteopenia alone in 10% of patients and thus masquerade as postmenopausal osteoporosis or senile osteoporosis, if monoclonal gammopathy and marrow plasmacytosis are not discovered [3,4]. Skeletal surveys remain negative in up to 20% of patients [3,4]. The increased sensitivity of computed tomography (CT) may identify focal lytic lesions of the spine in patients with back pain, increased plasma cells in the bone marrow, and a monoclonal protein in the serum but no radiologic changes of myeloma [25]. Magnetic resonance imaging may prove to be even more valuable for the early diagnosis of myeloma bone disease [26]. When lytic lesions are identified, they are most commonly found in the skull, ribs, clavicle, sternum, spine, pelvis, and proximal parts of the extremities [1,3]. The sharply demarcated, punched-out, sieve-like appearance of the calvarium is particularly well known. However, it should be realized that similar radiographic abnormalities may be caused by metastatic carcinoma, especially of the breast [1,3]. With vertebral collapse due to multiple myeloma, the intervertebral disk usually remains intact [l] and the vertebral pedicles are rarely involved, reflecting the relative paucity of red marrow in the pedicles as compared with the vertebral body [27]. Preservation of the pedicles, an intact intervertebral disk, and the more frequent presence of a paraspinal mass are guides that help distinguish multiple myeloma from metastatic car592
December 1992 The American Journal of Medicine
cinoma [1,3]. Because of the defective osteoblast response to the osteolysis, the lytic lesions receive poor uptake of radionuclides [28]. For this reason, bone scans generally underestimate the number of lytic lesions in patients with multiple myeloma. Osteosclerotic lesions occur in only about 3% of patients but can result in positive bone scans [29]. Osteosclerotic myeloma may be associated with chronic progressive polyneuropathy, hepatosplenomegaly, diabetes mellitus, and gynecomastia [30]. A single, particularly painful bone lesion can be palliated with moderate doses of local radiotherapy. Radiotherapy must, however, be limited to a small field and adjunctive chemotherapy delayed until the hematologic impact of the radiation is clear. Lesions usually persist, and radiographic evidence of skeletal repair is rare. Regression of lytic lesions on serial lateral skull roentgenograms has, however, been demonstrated in 30% of patients who respond to melphalan with a reduction in myeloma protein production [31]. Consequently, many attempts have been made to develop adjunctive therapy to treat or prevent widespread bone disease in multiple myeloma. In a large double-blind cooperative study, fluoride therapy was shown to be ineffective and possibly detrimental [32]. Another study examined a regimen of calcium carbonate combined with sodium fluoride, but there were no significant increases in bone density as measured by single-photon absorptiometry of the radial diaphysis, a cortical bone site, or of the distal radius, a primarily cancellous bone site [33]. Dichloromethylene bisphosphonate, a bone-seeking inhibitor of osteoclasts, decreases urinary calcium and hydroxyproline excretion in patients with active myeloma and, in some patients, has diminished bone pain [34,35]. This potentially useful drug is, however, not available in the United States. Even more promising are the reports of inhibition of osteolytic bone lesions with the more potent and longer-lasting agent, (3amino-1-hydroxypropylidene)1, l-bisphosphonate or APD, currently available as pamidronate disodium (Aredia, CIBA-GEIGY, Summit, NJ) [36,37]. In one report, the relief of bone pain with pamidronate lasted 10 to 24 months, but evidence of sclerosis within a previously lytic lesion occurred only once [38]. The side effects of this drug are confined to a transient increase in temperature of 1°C to 2OC and infrequent, mild, local irritation at the injection site. In short-term studies, salmon calcitonin has also been shown to decrease bone resorption and hypercalcemia in patients with multiple myeloma [39]. The hypocalcemic response to calcitonin may even be used to estimate the extent of lytic lesions [40]. Further studies are in progress to assess the value of pamidronate and calcitonin in
Volume 93
BONE INVOLVEMENT
the prevention and treatment of osteolytic lesions in multiple myeloma. Considerable interest has been focused on recalcification of lytic lesions as proof of a beneficial drug response, although this is not a particularly sensitive assessment of bone healing. Estimates of the amount of bone mineral that must be lost or gained to be evident to the trained radiologist range from 20% to 60%. Mariette and associates [5] analyzed bone density determinations in 10 patients with stage III myeloma. Stage III myeloma is characterized by a high tumor burden (greater than 1.2 X 1012 myeloma cells/m2) with a hemoglobin level less than 8.5 g/dL, a serum calcium level greater than 12 mg/dL, advanced lytic bone disease, or high M-component production rates [41]. The patients were treated with high-dose melphalan, carmustine, cyclophosphamide, and etoposide, total body irradiation, and autologous blood stem cell transplantation. Between stem cell collection by leukopheresis and autograft, the patients received three monthly courses of vincristine, Adriamycin, and methylprednisolone. Bone mineral density was measured at the lumbar spine and femoral neck by dual-energy radiograph absorptiometry (DXA or DEXA) at the time of diagnosis of the myeloma and 8 to 12 months after therapy. DEXA uses a stabilized radiographic tube to generate photons of two different energies (one photon is attenuated by bone, the other by soft tissue) and allows a much greater photon flux than with decay of nuclear sources. This state-of-the-art technique thus increases the image resolution and precision while minimizing the radiation exposure and time required for the measurements. Excluding one patient with profound initial osteopenia and an extraordinary gain in bone density after therapy, the remaining nine patients had a mean increase in lumbar density of 0.057 g/cm2, representing a gain of 7.7% (95% confidence interval, 2.2% to 12.2%). At the femoral neck, the mean difference was only -0.018 g/cm2 or -2.3%. Although the mean change at the femoral neck was not statistically significant, the 95% confidence interval was -7.3% to 2.8%, indicating that some patients lost a considerable amount of bone from the hip. Routine radiographic, CT, and magnetic resonance imaging examinations were done at the same intervals, and results did not noticeably change. With the DEXA procedure, Mariette and coworkers [5] found that the long-term coefficient of variation with a phantom of the lumbar spine was 0.5%. Short-term variability in four young volunteers with normal bone density was even better (0.4%) [42]. Even if these precision studies were done by the same technicians with the same densitometer, additional information on the clinical reliDecember
IN MULTIPLE
MYELOMA
/ WEINSTEIN
ability of the method is desirable. The patients with multiple myeloma were, after all, older and more osteopenic. In addition, the interval between the repeat measurements in the patients was about 1 year. A fairer comparison would involve the longterm reproducibility of the densitometric determinations in subjects of similar age and bone density as the experimental group. In the Metabolic Bone Disease Laboratory of the Medical College of Georgia, 56 white women, aged 45 to 84 years with a mean lumbar mineral density of 0.739 g/cm2, had repeat DEXA measurements made over a mean interval of 23 days (range: 2 to 68 days). The 95% confidence limit for the difference in the measurements of Ll-L4 was 0.032 g/cm2. Application of this rigorous test to the results of Mariette and associates [5] indicates that the increases in lumbar density found in the patients with multiple myeloma could be dependably detected by the DEXA technique. Finally, as to whether the increased density was due to the treatment regimen or the DEXA measurements, an instructive patient was recently seen at the Medical College of Georgia. A 53-year-old white man with multiple myeloma received conventional melphalan and prednisone therapy over a 9month interval. DEXA measurements of the lumbar spine increased from 0.724 to 0.790 g/cm2, representing a gain of 0.074 g/cm2 or 9.1%, whereas at the femoral neck, density decreased from 0.622 to 0.533 g/cm2, indicating a loss of 0.089 g/cm2 or 14.3% (95% confidence limit at the hip, 0.039 g/cm2). These findings are similar to those reported by Mariette et al [5] and suggest that the improvement in lumbar density is detectable because of the sensitivity of the new DEXA technique, rather than from more aggressive therapy of the myeloma. Furthermore, the simultaneous decrease in bone density at the femoral neck implies an increasing risk of hip fracture and stresses the vital importance of performing bone densitometry at multiple skeletal sites. The axial and appendicular bones respond differently to various systemic conditions and drug treatments. Augmentation of bone density in response to therapy is usually more conspicuous in the spine than in the appendicular skeleton [43]. The last sentence in the paper by Mariette and associates [5] is possibly the most important. Bone densitometry could well become a routinely measured prognostic factor in the treatment of patients with multiple myeloma. Now that the bone involvement can be precisely assessed with DEXA and promising therapeutic options with calcitonin or pamidronate are available [36-391, prevention of skeletal pain and fracture in multiple myeloma should no longer be considered impossible. 1992
The American
Journal
of Medicine
Volume
93
593
BONE INVOLVEMENT
IN MULTIPLE
MYELOMA
/ WEINSTEIN
REFERENCES 1. Snapper I. Bone diseases in medical practice.
New York: Grune & Stratton,
1957: 157-68. 2. Gregory WM. Richards MA. Malpas JS. Combination chemotherapy versus melphalan and prednisone in the treatment of multiple myeloma: an overview of published trials. J Clin Oncol 1992; 10: 334-42. 3. Kyle RA. Multiple myeloma: review of 869 cases. Mayo Clin Proc 1975; 50:
23. Warne GL. Fairley KF, Hobbs JB. Martin FIR. Cyclophosphamide-induced ovarian failure. N Engi J Med 1973; 289: 1159-62. 24. Gebharl SSP. Watts NB, Clark RV. Umpierrez G, Sgoutas D. Reversible impairment of gonadotropin secretion in critical illness. Arch Intern Med 1989;
149: 1637-41. 25. Soloman A, Rahamani R. Seligsohn U, Ben-Artzi involvement assessed by computerised tomography.
F. Multiple myeloma: early Skeletal Radio1 1984; 11:
29-40.
254-61.
4. Kapadia SB. Multiple myeloma: a clinicopathologic study of 62 consecutively autopsied cases. Medicine (Baltimore) 1980; 59: 380-92. 6. Mariette X, Khalifa P. Ravaud P, et a/. Bone densitometry in patients with multiple myeioma. Am J Med 1992; 93: 595-8. 6. Mundy GR, lbbotson KJ. D’Souza SM. Simpson EL, Jacobs JW. Martin TJ. The hypercalcemia of cancer. N Engl J Med 1984; 310: 1718-28. 7. Stashenko P. Dewhirst FE, Peros WJ. Kent RL, Ago JA. Synergistic interactions between interleukin 1. tumor necrosis factor, and lymphotoxin in bone resorption. J lmmunol 1987; 138: 1464-8.
26. Ludwig L. Tscholakoff D, Neuhold A, Fruhwald F. Rasoul S, Fritz E. Magnetic resonance imaging of the spine in multiple myeloma. Lancet 1987; 2: 364-6. 27. Jacobson HG. Poppel MH, Shapiro JH. Grossberger S. The vertebral pedicle sign: a roentgen finding to differentiate metastatic carcinoma from multiple myeloma. AIR Am J Roentgen01 1958; 801 817-21. 28. Wahner HW. Kyle RA. Beabout JW. Scintigraphic evaluation of the skeleton in multiple myeloma. Mayo Clin Proc 1980; 55: 739-46. 29. Frame B, Honasoge M. Kottamasu SR. Osteosclerosis. hyperostosis and related disorders. New York: Elsevier, 1987: 7&3. 30. lmawari M. Akatsuka N. lshibashi M. Beppu H, Suzuki H, Yoshitoshi Y. Syndrome of plasma cell dyscrasia, polyneuropathy and endocrine disturbance. Ann Intern Med 1974; 81: 490-3. 31. Rodriguez LH. Finkelstein JB, Shullenberger CC, Alexanian R. Bone healing in multiple myeloma with melphalan chemotherapy. Ann Intern Med 1972; 76:
6. Garrett IR, Durie BGM. Nedwin GE, eta/. Production of lymphotoxin, a boneresorbingcytokine, by cultured human myeloma cells. N Engl J Med 1987; 317:
526-32. 9. Bertolini DR. Nedwin GE, Bringman TS, Smith DD, Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumor necrosis factors. Nature 1986; 319: 516-8. 10. Stashenko P. Dewhirst FE, Rooney ML, Desjardins LA, Heeley JD. Interleukin-la is a potent inhibitor of bone formation in vitro. J Bone Miner Res 1987; 2:
559-65. 11. Durie BGM, Salmon SE, Mundy GR. Relation of osteoclast production
to extent of bone disease in multiple myeloma.
activating factor Br J Haematol 1981;
47: 21-30. 12. ValentinQpran tive histology
A, Charhon SA. Meunier PJ, Edouard CM. Arlot ME. Quantitaof myelomainduced bone changes. Br J Haematol 1982; 52:
601-10. 13. Niell HB. Neely CL, Palmieri GM. The postabsorptive
551-6. 32. Harley JB. Schilling A, Glidewell 0. Ineffectiveness of fluoride therapy in multiple myeloma. N Engl J Med 1975; 286: 1283-8. 33. Kyle RA. Jowsey J, Kelly PJ. Taves DR. Multiple myeloma bone disease. N Engl J Med 1975; 293: 1334-8. 34. Siris ES, Sherman WH, Baquiran DC, Schlatterer JP, Osserman EF, Canfield RE. Effects of dichloromethylene diphosphonate on skeletal mobilization of calcium in multiple myeloma. N Engl J Med 1980; 302: 310-5. 35. Delmas PD. Charhon S, Chapuy MC, et a/. Long-term effects of dichloromethylene diphosphonate (ClzMDP) on skeletal lesions in multiple myeloma. Metab Bone Dis Rel Res 1982; 4: 163-8. 36. van Breukelen FJM, Bijvoet OLM, van Oosterom AT. Inhibition of osteolytic bone lesions by (5amino-1-hydroxypropylidene)-1, I-bisphosphonate (A.P.D.). Lancet 1979; 1: 803-5. 37. Sleeboom HP, van Oosterom AT, Bijvoet OLM, Gleed JH, O’Riordan JLH. Comparison of intravenous (3-amino-1-hydroxypropylidene)-1. l-bisphosphonate and volume repletion in tumor-induced hypercalcemia. Lancet 1983; 2:
urinary hydroxyproline (spot-hypro) in patients with multiple myeloma. Cancer 1981; 48: 783-7. 14. Carlson K, Ljunghall S. Simonsson B, Smedmyr B. Serum osteocalcin concentrations in patients with multiple myeloma-correlation with disease stage and survival. J Intern Med 1992; 231: 133-7. 15. Nussbaum SR. Zahradnik RJ, Lavigne JR, et a/. Highly sensitive two-site immunoradiometric assay of parathyrin. and its clinical utility in evaluating patients with hypercalcemia. Clin Chem 1987: 33: 1364-7. 16. Lawson-Matthew PJ. Clayton JC, Guilland-Cumming DF, et a/. Defective production of 1,25(OH)zDs in myeloma. In: Norman AW, Schaefer K. Grigoleit HG, Herrath Dv. editors. Vitamin D: a chemical, biochemical and clinical update. Berlin: Walter de Gruyter & Co, 1985: 891-2. 17. Measuring the PTH level [editorial]. Lancet 1988; 1: 94-5. 18. Chen Y-H, Magalhaes MC. Hypoalbuminemia in patients with multiple myeloma. Arch Intern Med 1990; 150: 605-10. 19. Duggan DB. Schattner A. Unusual manifestations of monoclonal gammopathies: autoimmune and idiopathic syndromes. Am J Med 1986; 81: 864-70. 26. Nora NA. Singer I. Interpretation of hyparcalcemia in a patient with endstage renal disease. Arch Intern Med 1992; 152: 1321-2. 21. Lukert BP, Raisz LG. Glucocorticoid-induced osteoporosis: pathogenesis and management. Ann Intern Med 1990; 112: 352-64. 22. Ma&dams MR, White RH, Chipps BE. Reduction of serum testosterone levels during chronic $ucocorticoid therapy. Ann Intern Med 1986; 104:
du contenu mineral osseux par radiographie digitale quantitative. Premiers resultats au niveau vertebral lombrire. Presse Med 1989; 18: 1062-5. 43. Need AC, Nordin BEC. Which bone to measure? Osteoporosis Int 1990; 1:
648-51.
3-6.
594
December
1992
The American
Journal
of Medicine
Volume
239-43. 36. Thiebaud D. Leyvraz S, von Fliedner V, et a/. Treatment of bone metastases from breast cancer and myeloma with pamidronate. Eur J Cancer 1991; 27:
3741. 36. Rico H, Hernandez ER. Diaz-Mediavilla J. Alvarez A, Martinez R. Espinos D. Treatment of multiple myeloma with nasal spray calcitonin: a histomorphometric and biochemical study. Bone Miner 1990; 8: 231-7. 46. Bataille R, Sany J. Clinical evaluation of myeloma osteoclastic bone lesions: II. Induced hypocalcemia test using salmon calcitonin. Metab Bone Dis Rel Res
1982; 4: 39-42. 41. Salmon SL. Multiple myeloma. In: Wyngaarden JB, Smith LH, editors. Cecil textbook of medicine. 16th ed. Philadelphia: WB Saunders, 1982: 965-70. 42. Braillon P, Duboeuf F, Meaty MF, Barret P, Delmas PD. Meunier PJ. Mesure
93
