Hypercalcemia associated with head and neck malignancy is not an uncommon occurrence; its causes are multiple. Eight hypercalcemic patients with head and neck malignancy were studied. Serum calcium, serum phosphorus, tubular phosphorus threshold, fasting calcium ex- cretion, plasma 1,25-dihydroxyvitamin D, nephrogenous cyclic adeno- sine monophosphate (AMP), and immunoreactive parathyroid hor- mone were measured. Excessive dietary calcium administration in the form of an oral hyperalimentation preparation appeared to be the cause of hypercalcemia in 2 patients. Six patients demonstrated humorally mediated hypercalcemia. These patients resembled patients with primary hyperparathyroidism in having elevated nephrogenous cy- clic AMP excretion and reduced proximal tubular phosphorus reab- sorption, but they differed from patients with primary hyperparathyroid- ism by having normal levels of immunoreactive parathyroid hormone, markedly increased fasting calcium excretion, and strikingly reduced mean plasma levels of 1,25dihydroxyvitamin D. These data strongly suggest that the humoral factor responsible for hypercalcemia in pa- tients with head and neck cancer is not parathyroid hormone, and that patients with hyperparathyroidism can now be distinguished with confidence from those with malignancy-associated hypercalcemia.

HEAD & NECK SURGERY 5~125-129 1982

WITH HEAD AND NECK CANCER

MICHAEL F. ANGEL, MD, ANDREW STEWART, MD, MYLES L. PENSAK, MD, HAROLD R. C. PILLSBURY, MD, and CLARENCE T. SASAKI, MD

Hypercalcemia has been noted since the early 1900s t o be associated with malignancy. Al- though in early studies bone resorption by direct tumor invasion was felt to be the primary mecha- nism,’ more recent series have documented the presence of a humoral mechanism in some pa- tients. Snedecor and Baker2 coined the term “pseudohyperparathyroidism” (PHPT) to describe malignancy-associated hypercalcemia and hy- pophosphatemia occurring in the absence of bone metastases. Ariyan and associates3 examined hy- percalcemia in head and neck cancer patients and found that PHPT appeared to be the major mech- anism. Several other reports have appeared in the

From the Department of Surgery, Section of Otolaryngology (Drs Angel, Pensak, Pillsbury. and Sasaki), and the Department of Medicine, Division of Endocrinology (Dr. Stewart), Yale University School of Medicine. New Haven, CT.

This work was supported in part by NIH Grant RR125 and by the Adult General Clinical Research Center of the Yale-New Haven Hospital.

Address reprint requests to Dr. Angel at the Section of Otolaryngology, Yale School of Medicine, P . 0 Box 3333, 333 Cedar Street, New Haven, CT 0651 0.

Accepted for publication December 3, 1981

0 1982 John Wiley & Sons, Inc 01 48-640310502101 25 $01 2510

head and n.eck literature examining causes of hy- percalcemia and attempting to further define the responsible ~ubs tance .~ -~ Recent technical ad- vances, including measurement of nephrogenous cyclic adenosine monophosphate (AMP) and vita- min D metabolites, have allowed for more de- tailed biochemical evaluation of calcium metabo- lism in these patients.

We have recently evaluated a group of 50 con- secutive patients with malignancy-associated hypercalcemia, examining nephrogenous cyclic AMP excretion, plasma vitamin D metabolites, immunoreactive parathyroid hormone values, and renal calcium and phosphorus handling.7 In the course of this study we were struck by the fre- quency with which patients with head and neck cancer developed hypercalcemia. While a hu- moral factor appeared to be responsible for hyper- calcemia in all of these patients as previously r e p ~ r t e d , ~ in no patient could hypercalcemia be ascribed to parathyroid hormone secretion. Two patients in whom hypercalcemia was ini- tially attributed to head and neck cancer later proved to be due to excessive dietary calcium ad- ministration. This report describes these pa- tients in detail.

Cancer Hypercalcemia HEAD & NECK SURGERY NoviDec 1982 125

METHODS

Patients. From August 1978 through July 1980 the serum calcium results were reviewed on a daily basis in the clinical chemistry laboratories of the Yale-New Haven Hospital and the West Haven Veterans Administration Medical Center. The records of all patients whose serum calcium was above 11 mg/dl were examined; 65 patients with hypercalcemia and documented malignancy were so identified. Of the 65 patients, 8 had head and neck malignancy, and these cases form the basis of this report.

Samples. Spot urine and blood samples were ob- tained in the fasting (at least 8 hours) state, prior to the initiation of therapy for the tumor or for hypercalcemia. Specimens were analyzed and values calculated for total serum calcium (S Ca), serum phosphorus ( S PO4), fasting calcium excre- tion (F Ca E), renal phosphorus threshold (Tmp/ GFR), nephrogenous cyclic AMP (NcAMP), 25- hydroxyvitamin D [25(OH)Dl, 1,25-dihydroxy- vitamin D3 [ 1,25(OH)2Dl, and immunoreactive parathyroid hormone (iPTH).

Analyses. Serum calcium was measured by atomic absorption spectrophotometry. Serum and urine phosphorus were measured by an auto- mated modification of a standard method.8 Plasma and urine cyclic AMP were measured and nephrogenous cyclic AMP was calculated as pre- viously r e p ~ r t e d . ~ Plasma 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D3 were measured, using a modification" of the protein binding tech- nique." The parathyroid hormone radioimmuno- assay was performed at the Nichols Institute for Endocrinology in San Pedro, California.12 Fasting calcium excretion was calculated using spot urine calcium and creatinine concentration and simul- taneous serum creatinine. Renal phosphorus threshold was derived from a n0m0gram.l~ The bone scans were performed using Tc 99m meth- anediophosphorate. Staging of the tumors was performed using the TNM (tumor, node metas- tases) classification as revised by the American Joint Committee on Cancer.14 Partial data from some of these patients have been previously r ep~r t ed .~

RESULTS

Eight patients with squamous cell head and neck cancer and hypercalcemia were studied. Of these, 2 patients were receiving large amounts of cal-

cium (2,500 mglday and 3,000 mglday, respec- tively) in the form of oral hyperalimentation preparation (Isocal, Mead Johnson Nutritional, Evansville, ID). When oral hyperalimentation was discontinued, hypercalcemia disappeared, suggesting that the hypercalcemia in these pa- tients was dietary in nature.

Six patients were found to be hypercalcemic on the basis of their head and neck malignancies, and they accounted for 9% (6 of 65) of the cases of malignancy-associated hypercalcemia encoun- tered at the two hospitals during this study.

The clinical findings in the 6 patients are shown in Table 1. The primary locations of tumor are representative of those seen in patients with head and neck cancer. Since an equal number of poorly differentiated and well-differentiated tumors were encountered, the degree of differ- entiation did not appear to have an important bearing on the development of hypercalcemia. Hypercalcemia was a grave prognostic sign, with all 6 patients expiring within 6 months of its onset (mean = 3.3 months). The bone scan was negative in patient 6. Patients 2 and 5 each dis- played a single area of uptake. The scan displayed extensive uptake in patient 3. Scans were not per- formed on 2 patients. It was impossible to quanti- tate the tumor burden in each patient at the time of the onset of hypercalcemia. Ultimately 3 patients came to autopsy; all 3 had extensive met- astatic disease. Unfortunately the 3 remaining patients did not have autopsies, and thus one can only speculate about the amount of tumor burden in them.

The biochemical findings for the 6 patients are shown in Table 2. Among the 6 patients, serum phosphorus was normal in 3 and was reduced in 3. In contrast, the renal phosphorus threshold was strikingly reduced in all 6 patients, indicating that the patients were more phosphaturic than their serum phosphorus values might have sug- gested. Fasting calcium excretion was markedly elevated in all patients, on the average three times higher than is customarily seen in primary hyperparathyroidism. The 25(OH)D levels were normal, indicating the presence of adequate vita- min D stores. The 1,25(OHI2D values were extremely low in 3 patients, low in 2 patients, and elevated in patient 1. Nephrogenous cyclic AMP excretion was elevated in each patient and in the group as a whole, displaying values higher on the mean than, those usually encountered in primary hyperparathyroidism. Immunoreactive parathy- roid hormone values were normal in each patient.

126 Cancer Hypercalcernia HEAD & NECK SURGERY NoviDec 1982

Table 1. Clinical findings.

Survival time Stage of after onset

Age (yr)/ Primary presenta- Bone of hyper- Patient Sex location tation Histology scan Treatment' calcernia Autopsy findings

67 M

54 M

56 M

39 F

58 F

52 F

Floor of mouth

Floor of mouth

Tongue

Tongue

Tonsil

Larynx

Poorly differentiated squamous cancer

Poorly differentiated squamous cancer

We1 I-differentiated sauamous cancer

Poorly differentiated squamous cancer

Well-differentiated squamous cancer

Poorly differentiated squamous cancer

Not obtained

Positive left ramusi mandible

Positive right tibia, skull, ribs femur humerus

Not obtained

Positive left rib

Negative

Right hemiglos- sectorny, right R N D , radiation therapy

Left partial herniman- di bu lectomy; left R N D

Radiation therapy

Radiation therapy

Left partial glossectomy, left hemirnan- dibulectomy left FND, che- motherapy, radiation therapy

Laryngectomy, left R N D , radiation theraw

2 months

3 months

2 months

6 months

5 months

2 months

Cancer of tongue with local extension

Multiple right neck nodes

Multiple right su- praclavicular nodes

Metastases to lungs, pleura. Deritoneum

Metastases to heart, kidney, thyroid adrenals, pleura, peritoneum

N o t obtained

Not obtained

Not obtained

~

*FND = functional neck dissection; RND = radical neck dissection

Table 2. Biochemical findinas."

Fasting Serum Serum Serum calc ium

ca lc ium phosphorus creatinine TmpiGFR excretion 25(OH)D 1,25(OH),D NcAMP iPTH Patient (mgidl) (mg/dl) (mgidl GF) (mgidl GF) (mgidl GF) (ng/ml) (pgirnl) (nmolidl GF) (pl-Eq/ml)

1 16.1 2.7 1.3 1.5 0.74 32 22 4.87 100 2 12.7 2.3 0.6 1.6 0.45 16.5 41.7 4.2 55 3 13.5 3.6 1 .o 1.5 0.81 14 98 11.4 85 4 14.3 3.3 1 , I 2.0 0.87 25 7 6.23 76 5 11.9 2.1 0.6 1.8 0.55 16.1 41 3.89 24 6 15.8 3.5 1.8 1.4 2.06 10.5 7 5.05 66

Mean 14 2.9 1.07 1.6 0.91 19 36 5.94 56 Normal 8.5-10.6 3.1-4.5 0.6-1.5 2.7 0.18 15 22-64 0.5-2.5 100 Usual hy- 10.5-13.0 2.5-3.1 0.6-1.5 2.0-2.5 0.15-0.30 15 60-1 10 2.5-4.0 100-180

perparathy- ro id ranget

*iPTH = imrnunoreactive parathyroid hormone, NcAMP = nephrogenous cyclic adenosine monophosphate; 25(OHJD = 25-hydroxyvitarnin D, 1,25(OH),D = 1,25-dihydroxyvitamin D3, TmpiGFR = renal phosphorus threshold. +These are approximate or usual ranges based on the authors' experience.

Cancer Hypercal cem ia HEAD & NECK SURGERY Nov/Dec 1982 127

DISCUSSION

Hypercalcemia commonly accompanies squamous carcinomas of the head and neck. Ariyan and co- workers3 reported that 25% of patients with head and neck cancer will become hypercalcemic before their death. Hypercalcemia may result from one of several mechanisms in patients with malignant disease. Bone resorption by direct tumor invasion of the bone is the major mecha- nism of hypercalcemia in patients with breast cancer and multiple m y e l ~ m a . ' ~ , ~ ~ In other pa- tients, typically those with squamous cell carci- nomas, a humoral mechanism has been proposed. The precise nature of the humoral factor responsi- ble for this type of bone resorption remains elu- sive, with proposed candidates including prosta- g l a n d i n ~ , ' ~ ' ~ ~ parathyroid hormone,lg or a vitamin D-like sterol.

Farr and associates2' reviewed 100 patho- logically proven cases of primary hyperparathy- roidism; 34 of these patients developed a cancer elsewhere, either coincidental with or subsequent to the parathyroid adenoma. Thus, in some pa- tients with cancer, primary hyperparathyroidism may be a cause of hypercalcemia.

With development of the first immunoassay for parathyroid hormone (PTH) by Tashjian and associates2' and with later refinements of the assay by Berson and co-workers,22 the search for elevated PTH levels in patients manifesting PHPT or ectopic hyperparathyroidism was begun. Sherwood and c o - ~ o r k e r s ~ ~ were able to detect iPPH in the tumors of 7 of 13 patients with typi- cal PHPT. Palmieri and associate^'^ found iPTH in 6 out of 7 tumor extracts studied. Whether these extracts contained native parathyroid hor- mone or PTH-like substance remains unclear. In contrast Powell and associates24 were unable to detect PTH immunoreactivity in the serum or the tumor tissue of those with typical PHPT. Prosta- glandins El and E2 in tissue culture have been shown to be potent stimulators of bone resorp- t i ~ n . ~ ~ Clinical studied7 have suggested that hypercalcemia in certain patients is due to increased prostaglandin secretion. Seyberth and associates," in studying humoral mechanisms found 2 groups of patients: Group 1 (prostaglan- din dependent) displayed elevated urinary excre- tion of PGE-M, a metabolite of prostaglandin E2 indicative of increased production of prostaglandin E. Hypercalcemia in these patients responded to prostaglandin synthetase inhibitors. This group had normal urinary cyclic AMP excretion. Group 2 (prostaglandin independent) did not respond to

inhibitors and were characterized by elevated cyclic AMP secretion and elevated serum chloride similar to that seen in hyperparathyroidism. At present the relationship between prostaglandins and nephrogenous cyclic AMP is unclear and fur- ther study must await improvements in tech- niques for measuring prostaglandins.

Our group has recently provided further insight into the nature of the humoral factor responsible for humorally mediated hypercal- ~ e m i a . ~ We have shown that the majority of patients with malignancy-associated hypercal- cemia displayed evidence of tumor secretion of a calcemic, renotropic factor that resembled para- thyroid hormone in its ability to stimulate the proximal tubular PTH-sensitive adenylate cy- clase (resulting in increased nephrogenous cyclic AMP excretion) and to inhibit proximal tubular phosphate reabsorption. However, this factor dif- fered from PTH in its relative inability to stimu- late distal tubular calcium reabsorption (permit- ting marked hypercalcemia to occur), to stimulate proximal tubular 25 hydroxyvitamin D 1 hydrox- ylase (the enzyme responsible for ~ , ~ E I ( O H ) ~ D synthesis), or to react with PTH antisera.

The patients in the current study fall into this group of patients with humorally mediated hypercalcemia. Renal phosphorus thresholds were markedly reduced. Fasting calcium excre- tion far exceeded that customarily encountered, even in severe hyperparathyroidism. Plasma 1,25(OH)2D values were normal or low in all but 1 patient. Nephrogenous cyclic AMP excretion was markedly elevated, despite normal iPTH values in each patient. Using other PTH immu- noassays, we and others have demonstrated absent immun~reac t iv i ty .~ ,~~ The bone scans showed significant uptake in only 1 of 4 patients scanned.

The survival of these patients following onset of hypercalcemia was brief, a reminder of the advanced state of disease in patients who develop hypercalcemia from their malignancy. Inter- estingly, there appeared to be no import of degree of differentiation of tumor on the occurrence of hypercalcemia. To our knowledge this finding has not been commented on previously.

Two patients initially suspected of having malignancy-associated hypercalcemia proved to have diet-related hypercalcemia. Withdrawal of dietary calcium supplements resulted in a return to normal in serum calcium. These patients were receiving 2,500 to 3,000 mg of dietary calcium per day. Although this represents two to three times

128 Cancer Hypercalcemia HEAD & NECK SURGERY NoviDec 1982

the normal upper limits of calcium intake, it is surprising that it produced hypercalcemia since most patients with diet-related hypercalcemia (milk-alkali syndrome) have received substan- tially larger amounts of dietary ~ a l c i u m . ~ ~ , ~ ~ It is possible that other components in the diet (for example, lactose) enhanced the absorption of cal- cium in these patients.26 In any case, the impor- tant point to be made is that in any cancer patient, treatable causes of hypercalcemia due to factors other than malignancy may occur. These should be specificially sought and treated.

CONCLUSION

Hypercalcemia is not uncommon in patients with head and neck malignancy. There are several

REFERENCES

1. Gutman AB, Tyson LT, Gutman EB: Serum calcium inor- ganic phosphorus and phosphatase activity in hyper- parathyroidism, Paget’s disease, multiple myeloma, and neoplastic disease of the bones. Arch Intern Med 57:379- 413, 1936.

2. Snedecor PA, Baker HW: Pseudohyperparathyroidism due to malignant tumors. Cancer 17:1492, 1967.

3. Ariyan S, Farber LR, Hamilton BP, Papac RJ: Pseudohy- perparathyroidism in head and neck tumors. Cancer

4. Goodwin WJ, J r , Chandler JR: Hypercalcemia in epider- moid carcinoma of the head and neck. Am J Surg 132: 444-448, 1976.

5. Terz J J , Estep H, Bright R, Lawrence W, Curutchet HP, Kay S: Proceedings: Primary Oropharyngeal cancer and hypercalcemia. Cancer 33:334-339, 1974.

6. Liston, SL: Hypercalcemia and head and neck cancer. Arch Otolaryngol 104:597-600, 1978.

7. Stewart AF, Horst R, Deftos LJ, Cadman EC, Lang R, Broadus AE: Biochemical evaluation of patients with can- cer-associated hypercalcemia. N Engl J Med 303:1377- 1383, 1980.

8. Day JA, Ertingshausen G: Direct method for determining inorganic phosphate in serum with the “centrifichem.” Clin Chem 18:263-265, 1972.

9. Broadus AE: Nephrogenous cyclic AMP as a parathyroid function test. Nephron 23:136-141, 1979.

10. Horst RL, Shepard RM, Jorgenson WA, DeLuca HF: The determination of vitamin D metabolites on a single sam- ple: Changes during parturition in dairy cows. Arch Biochem Biophys 192:512-523, 1979.

11. Eisman JA, Hamstra AJ, Kream BE, DeLuca HF: A sen- sitive, precise, and convenient method for determination of 1,25-dihydroxyvitamin D in human plasma. Arch Biochem Biophys 176:235-243, 1976.

12. Nichols Institute: Parathyroid hormone. Nichols Institute Radioimmunoassay Manual. San Pedro, California, 1977,

33:159-166, 1974.

pp 105-112. 13. Biivoet. 0: Kidnev function in calcium and ohosahate

metabolism, in A;ioli, LV, Krane, SM (eds): ’Metabolic Bone Disease. New York, Academic Press, vol 1, 1977, pp

14. American Joint Committee on Cancer: Staging of Cancer 49-140.

potential causes, including dietary calcium excess, which is easily correctable. Bone metas- tasis is an uncommon cause; humoral mecha- nisms appear to be operative most frequently. Primary hyperparathyroidism, ectopic production of parathyroid hormone, and increased prosta- glandin synthesis have been proposed as humoral mechanisms of hypercalcemia. In head and neck cancer patients, elaboration of a parathyroid-like hormone distinct from native parathyroid hor- mone appears to be the usual cause. Patients with this humorally mediated hypercalcemia can be readily distinguished from those with pri- mary hyperparathyroidism by their markedly increased calcium excretion, low 1,25-dihydroxy- vitamin D levels, and normal immunoreactive parathyroid hormone levels.

of Head and Neck Sites and of Melanoma, Chicago, 1980. 15. Elion G, Mundy G R Direct resorption of bone by human

breast cancer cells in vitro. Nature 276:726-728, 1978. 16. Mundy GR, Raisz LG, Cooper RA, Schechter GP, Salmon

SE: Evidence for the secretion of an osteoclast stimulating factor in myeloma. N Engl J Med 291:1041-1046, 1974.

17. Seyberth HW, Segre GV, Morgan JL, Sweetman BJ, Potts JT, Jr , Oates JA Prostaglandins as mediators of hyper- calcemia associated with certain types of cancer. N Engl J Med 293:1278-1288, 1979.

18. Seyberth HW, Segre GV, Hamet P, Sweetman BJ, Potts JT, Jr, Oates JA Characterization of the group of patients with the hypercalcemia of cancer who respond to treat- ment with prostaglandin synthesis inhibitors. Trans As- soc Am Physicians 8942-104, 1976.

19. Palmieri GM, Nordquist RE, Omenn GS: Immunochem- ical localization of parathyroid hormone in cancer tissue from patients with ectopic hyperparathyroidism. J Clin Invest 53:1726-1735, 1974.

20. Farr HW, Fahey TJ, J r , Nash AG, Farr CM: Primary hy- perparathyroidism and cancer. Am J Surg 126539-543, 1973.

21. Tashjian AH, Levine L, Munson L Production of an anti- body to bovine parathyroid. Biochem Biophys Res Com- mun 8:259-265, 1962.

22. Berson SA, Yalow RS, Aurbach GD, Potts JT, Jr: Im- munoassay of bovine and human parathyroid hormone. Proc Natl Acad Sci 49:613-617, 1963.

23. Shenvood LM, O’Riordan JLH, Aurbach GD, Potts JT: Production of parathyroid hormone by non-parathyroid tumor. J Clin Endocrinol Metab 27:140-146, 1967.

24. Powell D, Singer FR, Murray TM, Minkin C, Potts JT, Jr: Nonparathyroid humoral hypercalcemia in patients with neoplastic diseases. N Engl J Med 289:176-181, 1973.

25. Klein DC, Raisz LG: Prostaglandins: Stimulation of bone resorption in tissue culture. Endocrinology 86:1436-1440, 1970.

26. McMillan DE, Freeman RB: The milk-alkali syndrome: A study of the acute disorder with comments on the develop- ment of the chronic condition. Medicine 44:485-501,1965.

27. Adams ND, Gray RW, Lemann J, Jr: The effects of oral CaCo, loading and dietary calcium deprivation on plas- ma 1,25-dihydroxyvitamin D concentrations in healthy adults. J Clin Endocrinol Metab 48:1008-1016, 1979.

Cancer Hypercalcemia HEAD & NECK SURGERY NoviDec 1982 129