The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
Brief Report
PreviousPrevious
Volume 330:1645-1649 June 9, 1994 Number 23
NextNext

Inhibition of Renal Phosphate Transport by a Tumor Product in a Patient with Oncogenic Osteomalacia
Qiang Cai, Stephen F. Hodgson, Pai C. Kao, Vanda A. Lennon, George G. Klee, Alan R. Zinsmiester, and Rajiv Kumar

 

Commentary
-Letters
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
In tumor-induced osteomalacia, a rare syndrome characterized by hypophosphatemia, hyperphosphaturia, low plasma 1,25-dihydroxyvitamin D concentrations, and osteomalacia,1,2,3,4,5 all biochemical and pathological abnormalities disappear when the tumor is removed. Tumors associated with this syndrome are thought to secrete a substance that inhibits the renal tubular reabsorption of phosphate,1,2,3,4,5 but whether this factor interacts directly with renal tubular cells is not known. We investigated the ability of medium in which sclerosing hemangioma cells from a patient with oncogenic osteomalacia were cultured to alter sodium-dependent phosphate transport in opossum-kidney epithelial cells. We found that the medium inhibited phosphate transport, without increasing cellular concentrations of cyclic adenosine monophosphate (cAMP). The medium had parathyroid hormone (PTH)-like immunoreactivity but no PTH-related protein immunoreactivity, and its action was not blocked by a PTH antagonist.

Case Report

The patient was first seen at the Mayo Clinic in November 1974, at the age of 47 years, with a seven-year history of aching that progressed through her arms and legs. She was the product of a normal pregnancy, and her growth and development had been normal. Physical examination revealed moderate proximal-muscle weakness. A movable, nontender mass measuring 2 by 1.5 cm was identified in the soft tissue of the distal anterior thigh. The serum calcium concentration was normal, the phosphate concentration was low, and the alkaline phosphatase concentration was high (Table 1). An iliac-bone biopsy done after double tetracycline labeling revealed osteomalacia. The patient was not treated with phosphate or vitamin D.

View this table:
[in this window]
[in a new window]
 
Table 1. Serum, Plasma, and Urinary Biochemical Values in a Patient with Oncogenic Osteomalacia.

 
The patient had a spontaneous fracture of her left proximal femur in September 1975. At that time, the subcutaneous mass in the distal thigh, which had enlarged, was excised. Histologic examination revealed a sclerosing hemangioma. The patient's fracture healed, her weakness improved, and her serum phosphate concentration increased to normal. She was well for the next 16 years. In October 1991, she began to have vague lower-extremity pain and noticed a mass at the site from which the hemangioma had been excised. Serum total and ionized calcium concentrations were normal, the phosphate concentration was low, and the alkaline phosphatase concentration was high (Table 1). The plasma concentration of 25-hydroxyvitamin D was normal and the concentration of 1,25-dihydroxyvitamin D was at the lower limit of the normal range6. Serum concentrations of sodium, potassium, glucose, total protein, aspartate aminotransferase, bilirubin, uric acid, creatinine, and albumin were normal. Urinalysis revealed osmolality of 489 mOsm per kilogram, with a pH of 5.3, a glucose concentration of 29 mg per deciliter, and no abnormal findings on microscopy. The pH of venous blood was 7.39.

By December 1992 the patient's movements and gait had become cautious, and she used her arms to rise from a sitting position. A firm, movable mass measuring 2 cm in diameter was present in the soft tissue of the distal anterior thigh beneath the healed surgical scar. Radiographic examination demonstrated degenerative arthritis of both hips and anterior compression of the T6 and T7 vertebral bodies. The serum or plasma concentrations of phosphate, calcium, 1,25-dihydroxyvitamin D, and osteocalcin7 were low (Table 1), as was the maximal capacity of the renal tubules to reabsorb phosphorus divided by the glomerular filtration rate8. Serum or plasma concentrations of PTH,9 PTH-related protein,10 25-hydroxyvitamin D, and alkaline phosphatase were normal. A transiliac-bone biopsy, performed after double tetracycline labeling, showed a mineralization defect (osteomalacia) (Figure 1A and Figure 1B)11.The soft-tissue mass was excised in January 1993; pathological examination again revealed a sclerosing hemangioma (Figure 1C). Postoperatively, when the patient was asymptomatic, her serum phosphate and plasma 1,25-dihydroxyvitamin D concentrations were normal, but the serum osteocalcin concentration was elevated (Table 1).


View larger version (28K):
[in this window]
[in a new window]
 
Figure 1. Histologic Sections of Bone and Tumor Tissue from a Patient with Oncogenic Osteomalacia.

In Panel A, undecalcified iliac trabecular bone (stained green) is covered by excessive amounts of osteoid (stained orange-red, arrows), especially in the center of the specimen (Goldner, x6). In Panel B, fluorescence microscopy of bone after double tetracycline labeling in vivo reveals only single-fluorescent yellow labels (arrows) instead of discrete double-fluorescent labels (x6). These findings are consistent with the presence of a mineralization defect (osteomalacia). In Panel C, histopathological analysis of tumor tissue removed from the patient in 1993 shows a typical sclerosing hemangioma (hematoxylin and eosin, x260).

 
Methods

Serum, plasma, and urinary constituents were measured by standard techniques in the Mayo Medical Laboratories.

Cell Culture

For the tumor-cell culture, 1.5 g of the tumor tissue removed in 1993 was dissociated enzymatically in phosphate-buffered saline containing 0.8 percent collagenase, 0.25 percent trypsin, and 0.02 mg of DNAase per milliliter. The cells were washed once in RPMI medium and resuspended in 30 ml of RPMI medium containing 10 percent calf bovine serum. The cells were plated into three tissue-culture dishes at a concentration of 1.3 x 105 cells per milliliter and cultured at 37 °C in a humidified atmosphere of 95 percent air and 5 percent carbon dioxide. The cells grew slowly in vitro. Conditioned medium was removed on days 2 and 8, pooled, and stored at -70 °C.

Opossum-kidney cells (a gift of Dr. Leonard Forte, University of Missouri) were cultured in 45 percent Dulbecco's modified Eagle's medium and 45 percent F12 medium, with 10 percent fetal-calf serum, 100 U of penicillin per milliliter, and 100 µg of streptomycin per milliliter. The cells then were seeded at a density of about 1 x 105 per well in 24-well tissue-culture plates. Assays were carried out two to three days after the cells were confluent.

Measurement of Sodium-Dependent Phosphate, Alanine, and Glucose Cotransport

The method used to measure sodium-dependent phosphate, alanine, and glucose cotransport has been described in detail elsewhere12,13. In brief, 100 microl of growth medium obtained from the tumor-cell cultures on day 2, 100 microl of RPMI medium with 10 percent fetal-calf serum, bovine PTH 1-34 (6 x 10-9 M), or PTH vehicle was added to the opossum-kidney cells to assess their effect on sodium-dependent phosphate transport. [Nle8, Nle18, Tyr34] bovine PTH (3-34) amide (10-5 M) was used as a PTH antagonist.

For the measurement of sodium-dependent phosphate transport, 0.1 mM dibasic potassium phosphate was included in the transport medium and [32P]dibasic potassium phosphate was added to a final specific activity of 2 micro Ci per milliliter. For sodium-dependent alanine transport, 0.1 mM l-alanine and [3H]l-alanine were added (final specific activity, 1 micro Ci per milliliter). For glucose transport, 0.1 mM methyl-{alpha}-glucopyranoside and methyl(alpha-d-[u-14C] gluco) pyranoside were added (final specific activity, 0.2 micro Ci per milliliter). The transport of phosphate, alanine, and methyl-alpha-glucopyranoside was assayed separately. Each transport reaction was measured in three or four duplicate wells. Each assay included three or four blank wells to correct for solute bound to the surfaces of the cells and the well and in intercellular spaces.

Treatment of Medium Obtained from Tumor-Cell Cultures

Medium (0.5 ml) obtained on day 2 of tumor-cell culture and control medium were boiled for 10 minutes in a water bath. Medium (0.5 ml) harvested from the tumor-cell cultures was dialyzed against water at 4 °C overnight with membranes that retained particles with a molecular weight of either 8 kd or larger or 25 kd or larger. The effect of the boiled or dialyzed medium on phosphate transport in opossum-kidney cells was measured.

Assays

Enzyme immunoassay was used to measure cAMP14. PTH-like substances were sought in medium obtained from tumor-cell cultures on day 2 with an immunochemiluminometric midregion assay9. This assay measures intact PTH and N-terminal PTH fragments but not C-terminal PTH (53-84). RPMI medium containing either 10 percent fetal-calf serum or 10 percent fetal-calf serum conditioned by the growth of Chinese hamster-ovary cells was used as the control medium. PTH-related protein was measured by radioimmunoassay with antibodies against PTH-related protein10.

Tumor Implantation into Nude Mice

Small fragments of the excised tumor were implanted into five nude mice. Serum phosphate then was measured periodically in the mice.

Statistical Analysis

The results of the studies of solute uptake by opossum-kidney cells were analyzed by analysis of variance for a randomized block design (trays being blocks). An overall test for treatment effects in this analysis was therefore adjusted for differences between trays. Pairwise comparisons (tumor medium vs. vehicle and PTH vs. PTH vehicle) were done at an alpha level of 0.025 to adjust for two pairwise comparisons. The effect of dialysis was examined by comparing tumor medium with control medium before and after dialysis with membranes capable of retaining particles 8 kd or larger or particles 25 kd or larger, with a two-sample t-test at an alpha level of 0.017 (i.e., adjustment for three tests). All statistical tests were two-sided.

Results

Histologic and histomorphometric analysis of the transiliac-bone specimen obtained by biopsy in 1993 showed a mineralization defect (osteomalacia) (Figure 1A and Figure 1B). The tumor (Figure 1C) was a sclerosing hemangioma.

The effect of the medium obtained from tumor-cell cultures on sodium-dependent phosphate transport in opossum-kidney cells was compared with that of bovine PTH 1-34. Medium obtained on day 2 of tumor-cell culture significantly inhibited phosphate transport in cultured opossum-kidney cells (Figure 2), as did medium harvested from the tumor cells on day 8 (data not shown). Medium collected on or after day 15 did not inhibit phosphate transport. Medium obtained on day 2 did not significantly inhibit alanine and glucose transport (Figure 2).


View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Effect of Medium Obtained from Tumor-Cell Cultures on Solute Transport in Cultured Epithelial Opossum-Kidney Cells.

In the top panel, 6 x 10-9 M bovine PTH, 100 microl of medium obtained from tumor-cell cultures, 100 microl of RPMI medium, and 100 microl of RPMI medium from cultures of Chinese hamster-ovary (CHO) cells were used. In the bottom panel, 100 microl of each medium was used. RPMI medium was used as a control. Values are the means (±SE) of three to five experiments, with three to four replicate samples analyzed per experiment.

 
Boiling the medium obtained from tumor-cell cultures on day 2 for 10 minutes abolished the inhibition of phosphate transport in opossum-kidney cells. Medium dialyzed with a membrane capable of retaining particles 8 kd or larger inhibited sodium-dependent phosphate transport in opossum-kidney cells, whereas medium dialyzed with a membrane capable of retaining particles 25 kd or larger did not (Table 2). [Nle8, Nle18, Tyr34] bovine PTH (3-34) amide did not block the phosphate-inhibiting effect of medium obtained from tumor-cell cultures in opossum-kidney cells.

View this table:
[in this window]
[in a new window]
 
Table 2. Effect of Dialysis on the Inhibitory Properties of Medium Obtained from Tumor-Cell Cultures on Phosphate Transport in Opossum-Kidney Cells.

 
The effect of the medium obtained from tumor-cell cultures on the production of cAMP was compared with that of the bovine PTH 1-34. As shown in Figure 3, bovine PTH 1-34 increased the accumulation of cAMP in opossum-kidney cells, whereas medium from tumor-cell cultures did not.


View larger version (10K):
[in this window]
[in a new window]
 
Figure 3. Effect of Medium Obtained from Tumor-Cell Cultures on cAMP Accumulation in Cultured Epithelial Opossum-Kidney Cells.

In each analysis, 6 x 10-9 M PTH, 100 microl of medium obtained from tumor-cell cultures, and 100 microl of RPMI medium were used. RPMI medium was used as a control. Values are the means (±SE) of three experiments with three to four replicate samples analyzed per experiment.

 
PTH-like immunoreactivity was twice as high in medium obtained from tumor-cell cultures on day 2 as in control RPMI medium or RPMI medium exposed to Chinese hamster-ovary cells (2.1 pmol per liter vs. 0.9 and 1.2 pmol per liter, respectively). When the medium from tumor-cell cultures was serially diluted with PTH-free serum, the resultant values differed from those that would have been obtained had authentic PTH or an N-terminal fragment of PTH been present in the sample (undiluted medium, 2.1 pmol per liter; 1:2 dilution, 0.4 pmol per liter; 1:4 dilution, 0.09 pmol per liter; and 1:8 dilution, 0.02 pmol per liter). PTH-related protein was not detectable in the medium obtained from tumor-cell cultures or in the patient's serum.

The mean (±SE) serum phosphate concentrations in the five nude mice 6, 9, and 10 months after the implantation of tumor fragments were 8.4 ±0.2 mg per deciliter (2.7 ±0.1 mmol per liter), 4.7 ±0.9 mg per deciliter (1.5 ±0.3 mmol per liter), and 5.4 ±0.2 mg per deciliter (1.5 ±0.1 mmol per liter), respectively. The corresponding values in two control nude mice were 9.3 ±2.4 mg per deciliter (3.0 ±0.7 mmol per liter), 6.2 ±0.6 mg per deciliter (2.0 ±0.2 mmol per liter), and 7.2 ±0.1 mg per deciliter (2.3 ±0.05 mmol per liter).

Discussion

Decreased renal phosphate transport is a rare cause of metabolic bone disease; such a defect occurs in Fanconi's syndrome, familial hypophosphatemia,1,2 and tumor-associated hypophosphatemia (oncogenic osteomalacia)15,16,17,18,19. Our patient had the typical manifestations of tumor-associated hypophosphatemia. Removal of the tumor was associated with reversal of the hypophosphatemia and all other metabolic abnormalities.

Cell-culture medium conditioned by the growth of tumor cells from the patient inhibited sodium-dependent phosphate transport in cultured renal cells. We did not measure sodium-independent phosphate transport because this contributes little (<6 percent) to total phosphate transport in the kidney under these experimental conditions12. The inhibitory effect was lost when the medium was boiled or was dialyzed against a membrane capable of filtering 25-kd particles. The inhibitory effect on phosphate transport in the cultured renal cells was independent of the accumulation of cAMP. We have no information regarding the nature of the second messenger (if any) that mediates the effect of the tumor factor on the kidney. The absence of an increase in cAMP in the renal cells after the addition of medium obtained from tumor-cell cultures argues against a role for PTH, PTH-related protein, or an N-terminal bioactive fragment of these hormones in the pathogenesis of hypophosphatemia. The absence of both hypercalcemia and increased serum PTH concentrations in the patient supports this contention. Furthermore, the finding that a PTH antagonist had no effect on the changes in phosphate transport in opossum-kidney cells induced by the medium from tumor-cell cultures suggests that the factor is not PTH itself.

Phosphaturia contributes to the pathogenesis of osteomalacia in patients with oncogenic osteomalacia. PTH or PTH-related peptide, well-known phosphaturic agents,20,21 cannot be the mediators of the phosphaturia because the patients have normal serum calcium and PTH concentrations. Urinary excretion of cAMP is not increased in patients with this syndrome, but it usually is increased in patients with hyperparathyroidism or cancer-associated hypercalcemia caused by PTH-related peptide. Serum phosphate concentrations in nude mice implanted with pieces of the tumor gradually decreased after implantation; however, the pattern of change in the control mice was similar. The clinical course of the patient after tumor removal is consistent with the presence of a circulating phosphaturic factor.

Plasma 1,25-dihydroxyvitamin D concentrations are low in patients with oncogenic osteomalacia, despite the presence of hypophosphatemia,21,22,23 which usually increases plasma 1,25-dihydroxyvitamin D concentrations by stimulating the renal 25-hydroxyvitamin D-1{alpha}-hydroxylase in a PTH-independent manner24. Deficient production of 1,25-dihydroxyvitamin D could be a factor contributing to the pathogenesis of oncogenic osteomalacia in these patients,4,25,26 because in some patients the clinical and biochemical abnormalities improve during calcitriol therapy1. In a recent study, 25-hydroxyvitamin D-1alpha-hydroxylase activity in cultured renal tubular cells was decreased by incubating the cells with tumor extracts; the extracts presumably contained a substance that inhibited the formation of 1,25-dihydroxyvitamin D26.

We have demonstrated that culture medium conditioned by the growth of sclerosing hemangioma cells obtained from a patient with oncogenic osteomalacia inhibited phosphate transport in renal epithelial cells. The factor elaborated by the tumor cells has a low molecular weight and is heat-sensitive.

Supported by grants (DK 25409, DK 42971, CA-37343, and RR 00585) from the National Institutes of Health.


Source Information

From the Nephrology Research Unit (Q.C., R.K.), the Neuroimmunology Research Laboratory (V.A.L.), the Section of Biostatistics (A.R.Z.), and the Departments of Medicine (S.F.H., R.K.), Neurology and Immunology (V.A.L.), and Laboratory Medicine and Pathology (P.C.K., G.G.K.), Mayo Clinic, Rochester, Minn.

Address reprint requests to Dr. Kumar at the Mayo Clinic, 200 First St., SW, 911A Guggenheim Bldg., Rochester, MN 55905.

References

  1. Lobaugh B, Burch WM Jr, Drezner MK. Abnormalities of vitamin D metabolism and action in the vitamin D resistant rachitic and osteomalacic diseases. In: Kumar R, ed. Vitamin D: basic and clinical aspects. Boston: Martinus Nijhoff, 1984:665-720. 
  2. Fanconi A, Fischer JA, Prader A. Serum parathyroid hormone concentrations in hypophosphataemic vitamin D resistant rickets. Helv Paediatr Acta 1974;29:187-194. [Medline]
  3. McCance RA. Osteomalacia with Looser's nodes (milkman's syndrome) due to a raised resistance to vitamin D acquired about the age of 15 years. Q J Med 1947;16:33-46. [Free Full Text]
  4. Fukumoto Y, Tarui S, Tsukiyama K, et al. Tumor-induced vitamin D-resistant hypophosphatemic osteomalacia associated with proximal renal tubular dysfunction and 1,25-dihydroxyvitamin D deficiency. J Clin Endocrinol Metab 1979;49:873-878. [Free Full Text]
  5. Weidner N, Bar RS, Weiss D, Strottmann MP. Neoplastic pathology of oncogenic osteomalacia/rickets. Cancer 1985;55:1691-1705. [CrossRef][Medline]
  6. Kao PC, Heser DW. Simultaneous determination of 25-hydroxy- and 1,25-dihydroxyvitamin D from a single sample by dual-cartridge extraction. Clin Chem 1984;30:56-61. [Free Full Text]
  7. Kao PC, Riggs BL, Schryver PG. Development and evaluation of an osteocalcin chemiluminoimmunoassay. Clin Chem 1993;39:1369-1374. [Abstract]
  8. Walton RJ, Bijvoet OLM. Nomogram for derivation of renal threshold phosphate concentration. Lancet 1975;2:309-310. [Medline]
  9. Klee GG, Preissner CM, Schryver PG, Taylor RL, Kao PC. Multisite immunochemiluminometric assay for simultaneously measuring whole-molecule and amino-terminal fragments of human parathyrin. Clin Chem 1992;38:628-635. [Free Full Text]
  10. Kao PC, Klee GG, Taylor RL, Heath H III. Parathyroid hormone-related peptide in plasma of patients with hypercalcemia and malignant lesions. Mayo Clin Proc 1990;65:1399-1407. [Medline]
  11. Hodgson SF, Johnson KA, Muhs JM, Lufkin EG, McCarthy JT. Outpatient percutaneous biopsy of the iliac crest: methods, morbidity, and patient acceptance. Mayo Clin Proc 1986;61:28-33. [Medline]
  12. Al-Mahrouq HA, Rassier ME, Dousa TP, Kempson SA. Sodium-dependent transport by cultured proximal tubule cells. J Tissue Cult Methods 1991;13:185-90.
  13. Smith PK, Krohn RI, Hermanson GT, et al. Measurement of protein using bicinchoninic acid. Anal Biochem 1985;150:76-85. [Erratum, Anal Biochem 1987;163:279.] [CrossRef][Medline]
  14. Horton JK, Martin RC, Kalinka S, et al. Enzyme immunoassays for the estimation of adenosine 3',5' cyclic monophosphate and guanosine 3',5' cyclic monophosphate in biological fluids. J Immunol Methods 1992;155:31-40. [CrossRef][Medline]
  15. Harrison HE. Oncogenous rickets: possible elaboration by a tumor of a humoral substance inhibiting tubular reabsorption of phosphate. Pediatrics 1973;52:432-434. [Free Full Text]
  16. Ryan EA, Reiss E. Oncogenous osteomalacia: review of the world literature of 42 cases and report of two new cases. Am J Med 1984;77:501-512. [CrossRef][Medline]
  17. Popovtzer MM. Tumor-induced hypophosphatemic osteomalacia (TUO): evidence for a phosphaturic cyclic AMP-independent action of tumor extract. Clin Res 1981;29:418A-418A.abstract 
  18. Aschinberg LC, Solomon LM, Zeis PM, Justice P, Rosenthal IM. Vitamin D-resistant rickets associated with epidermal nevus syndrome: demonstration of a phosphaturic substance in the dermal lesions. J Pediatr 1977;91:56-60. [CrossRef][Medline]
  19. Salassa RM, Jowsey J, Arnaud CD. Hypophosphatemic osteomalacia associated with "nonendocrine" tumors. N Engl J Med 1970;283:65-70.
  20. Caverzasio J, Rizzoli R, Bonjour JP. Sodium-dependent phosphate transport inhibited by parathyroid hormone and cyclic AMP stimulation in an opossum kidney cell line. J Biol Chem 1986;261:3233-3237. [Free Full Text]
  21. Weidner N. Review and update: oncogenic osteomalacia-rickets. Ultrastruct Pathol 1991;15:317-333. [Medline]
  22. Sweet RA, Males JL, Hamstra AJ, DeLuca HF. Vitamin D metabolite levels in oncogenic osteomalacia. Ann Intern Med 1980;93:279-280.
  23. Drezner MK, Feinglos MN. Osteomalacia due to 1{alpha},{alpha}25-dihydroxycholecalciferol deficiency: association with a giant cell tumor of bone. J Clin Invest 1977;60:1046-1053.
  24. Ribovich ML, DeLuca HF. Effect of dietary calcium and phosphorus on intestinal calcium absorption and vitamin D metabolism. Arch Biochem Biophys 1978;188:145-156. [CrossRef][Medline]
  25. Drezner MK, Lobaugh B, Lyles KW, Carey DE, Paulson DF, Harrelson JM. The pathogenesis and treatment of tumor-induced osteomalacia. In: Norman AW, Schaefer K, eds. Vitamin D: chemical, biochemical, and clinical endocrinology of calcium metabolism. Berlin, Germany: Walter de Gruyter, 1982:949-54.
  26. Miyauchi A, Fukase M, Tsutsumi M, Fujita T. Hemangiopericytoma-induced osteomalacia: tumor transplantation in nude mice causes hypophosphatemia and tumor extracts inhibit renal 25-hydroxyvitamin D1-hydroxylase activity. J Clin Endocrinol Metab 1988;67:46-53. [Free Full Text]

 

Commentary
-Letters
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Oncogenic Osteomalacia
Cotton G. E., Kumar R., Hodgson S. F.
Extract | Full Text  
N Engl J Med 1994; 331:1023, Oct 13, 1994. Correspondence

Computer-Based Diagnostic Systems
Lehmann E. D., Yolton R. L., Berner E. S., Webster G. D., Shugerman A. A.
Extract | Full Text  
N Engl J Med 1994; 331:1023-1024, Oct 13, 1994. Correspondence

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved.