- Email: [email protected]
Incidence of humoral hypercalcemia of malignancy among hypercalcemic patients with cancer
Clinica Chimica Acta 453 (2016) 190–193
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Incidence of humoral hypercalcemia of malignancy among hypercalcemic patients with cancer Jeffrey J. Szymanski 1, Zaher K. Otrock 1, Khushbu K. Patel, Mitchell G. Scott ⁎ Department of Pathology and Immunology, Washington University, St Louis, MO, United States
a r t i c l e
i n f o
Article history: Received 12 September 2015 Received in revised form 11 December 2015 Accepted 14 December 2015 Available online 17 December 2015 Keywords: Hypercalcemia Malignancy PTHrP Prevalence
a b s t r a c t Background: Malignancy-associated hypercalcemia (MAHC) is the most common cause of hypercalcemia among hospitalized patients. MAHC can result from the production of parathyroid hormone related peptide (PTHrP) which is known as humoral hypercalcemia of malignancy (HHM). HHM is commonly thought to account for approximately 80% of MAHC. Methods: We conducted a 12-year review of PTHrP testing at our institution to establish the prevalence of HHM among patients with MAHC. Results: A total of 524 PTHrP immunoassays were performed during the study period of which 470 tests qualified for inclusion in the analysis. Evidence of malignancy was found for 242 of 470 patients (51%). No etiology could be determined for 98 cases of MAHC (40%) and increased PTHrP contributed to 92 cases (38%) of MAHC. Age, race and gender were not associated with HHM. Increased PTHrP was observed at initial malignancy diagnosis in 20% of cases. PTHrP was never increased outside of the context of malignancy. Discussion: The prevalence of HHM among patients with MAHC is likely to be lower than previously described. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Malignancy-associated hypercalcemia (MAHC) occurs in 20–30% of cancer patients [1,2] and is the most common cause of hypercalcemia among hospitalized patients [3]. MAHC usually results from osteolytic bone lesions or from the production of parathyroid hormone related peptide (PTHrP), the latter of which is known as humoral hypercalcemia of malignancy (HHM). Nine of the 13 N-terminal amino acids of human PTHrP are homologous to those of parathyroid hormone, and PTHrP exerts its effects through binding to the parathyroid hormone 1 (PTH-1) receptor [4]. PTHrP is a peptide widely expressed throughout the body and circulates at low levels in the plasma of healthy individuals [5]. PTHrP is also ectopically secreted at much higher levels by some solid tumors and in some hematologic malignancies [6]. In these cancers, secreted PTHrP causes hypercalcemia via increased calcium absorption at the kidney and increased bone resorption [7]. Diagnostic workup of suspected HHM begins with the measurement of ionized calcium and PTH [1,8]. In patients with high ionized calcium and low PTH, measurement of PTHrP can confirm the diagnosis of HHM. HHM has been commonly thought to account for approximately 80% of MAHC [1,7]. This value is based on studies either predating the ⁎ Corresponding author at: 660 S. Euclid Ave, Campus Box 8118, St. Louis, MO 63110, United States. E-mail address: [email protected] (M.G. Scott). 1 JS and ZKO contributed equally to this manuscript.
http://dx.doi.org/10.1016/j.cca.2015.12.017 0009-8981/© 2015 Elsevier B.V. All rights reserved.
identification of PTHrP [9] or during validation of the original PTHrP immunoassay [10]. However, some early PTHrP immunoassay studies [11], and a recent, smaller study [8] reported lower prevalence of HHM. Our experience with PTHrP as a "resident-approval" test also challenged the accepted high prevalence of HHM in MAHC. 2. Methods 2.1. Study design To determine the prevalence of HHM in hospitalized patients with MAHC, we conducted a single-institution, cross sectional study of all PTHrP tests ordered at Barnes-Jewish Hospital in St. Louis over a twelve-year period from January 1, 2002 through December 31, 2013. PTHrP values were excluded from the study if they were not the first measurement for a given hospital admission, if they were performed on patients with normal or low calcium, or if associated patient records were unavailable. For each included PTHrP measurement, a chart review was performed to establish patient demographics, history of malignancy, relevant admitting laboratory values, etiology of hypercalcemia, and hospital admission outcome. This study was approved by the Washington University Institutional Review Board. 2.2. PTHrP immunoassays During the period of investigation, samples for PTHrP analysis were sent to one of four reference laboratories: Mayo Medical Laboratories
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193
191
Table 2 Demographics of patients with both elevated serum calcium and evidence of malignancy. Significance of age by Mann–Whitney U test. Significance of gender and race by Pearson chi-squared test. Age is in years. Gender and race values are total number of patients. HHM: humoral hypercalcemia of malignancy.
Fig. 1. Study design.
(311 assays), Quest Diagnostics (88 assays), Specialty Laboratories — Nichols Institute Diagnostics (30 assays), and ARUP Laboratories (41 assays). For all reference laboratories, PTHrP was determined using sandwich immunoassay. Results were evaluated against the reference interval provided by each laboratory and those exceeding the upper limit of the reference interval (1.9, 4.0, 4.7, or 1.3 pmol/l for the above mentioned reference laboratories, respectively) were considered increased. Positivity rates ranged from 7% to 20%. Specimens for PTHrP testing were collected according to the requirements given by each of the referral laboratories. 2.3. Clinical data collection PTHrP results and patient age and sex were collected from the hospital laboratory information system (Cerner Millennium). Patient race and admitting laboratory values were collected from the hospital electronic medical record (Clinical Desktop 2). Laboratory values included total and ionized calcium, albumin, PTH, phosphate, and 25-OH vitamin D. Total calcium (by colorimetric methods), total ionized calcium (by direct ion-selective electrode), albumin (by bromocresol green), PTH (by chemiluminescence enzyme immunoassay), phosphate (by ammonium molybdate), and 25-OH vitamin D (by immunoassay) were determined in-house. Length of stay and inpatient mortality were obtained from the hospital billing department. Type of malignancy and etiology of Table 1 Etiology of hypercalcemia among 242 patients with both elevated serum calcium and evidence of malignancy. Some patients had multiple causes of hypercalcemia. HHM: humoral hypercalcemia of malignancy. Etiology
Number of patients
Percent of patients
Undetermined HHM Osteolytic bone lesions Immobilization Primary hyperparathyroidism Tertiary hyperparathyroidism Dehydration Granulomatous disease Thiazide diuretics Calcium supplementation
98 92 66 8 6 2 2 2 1 1
40.5 38.0 27.3 3.3 2.5 0.8 0.8 0.8 0.4 0.4
HHM
Other etiology
p
Age (median)
59
60
0.528
Gender Male Female
55 36
93 58
0.859
Race Caucasian African American Other
59 31 1
98 51 2
0.96
hypercalcemia were determined by a manual review of laboratory values and patient charts by residents and fellows in the Department of Pathology and Immunology at Washington University. PTHrP was considered to contribute to MAHC if a patient had increased plasma total (N 10.3 mg/dl) or ionized (N5.1 mg/dl) calcium and an initial PTHrP value higher than the upper limit of the reference interval as provided by each of the reference laboratories. Osteolytic bone lesions were considered to contribute to MAHC if such lesions were documented on a radiographic imaging study or at autopsy. Other etiologies of hypercalcemia were identified through manual review of the electronic medical record. The etiology of a patient's hypercalcemia was considered unknown only if both PTH and PTHrP were normal and no cause was suggested or stated in pathology, radiology, or oncology notes or the discharge summary. 2.4. Data analysis All data analysis was performed in Microsoft Excel (ver 10, Microsoft) and SPSS Statistics (ver 22, IBM). Categorical variables were compared with Pearson chi-square or Fisher exact tests as appropriate. Association of individual laboratory values with etiology of hypercalcemia was assessed by Mann–Whitney U test. Laboratory results are presented as median values unless otherwise stated. 3. Results 3.1. PTHrP immunoassays A total of 524 PTHrP immunoassays were performed through Barnes-Jewish Hospital from January 1, 2002 to December 31, 2013 (Fig. 1). Fifty-four PTHrP tests (10%) did not meet inclusion criteria as Table 3 Site of primary malignancy among patients with both elevated serum calcium and evidence of malignancy. Three patients had multiple, concurrent cancer types. Significance by Fisher exact test or (*) for Pearson chi-squared test. HHM: humoral hypercalcemia of malignancy. Site of primary malignancy
Total
HHM
Other etiology
p
Head and neck Lung Leukemia or lymphoma Renal Breast Multiple myeloma Hepatocellular carcinoma Bladder Colon Pancreatic Thyroid Cholangiocarcinoma Ovarian Other
61 46 23 21 17 14 11 8 8 4 4 4 3 21
27 22 4 6 5 0 4 6 4 1 0 4 2 7
34 24 19 15 12 14 7 2 4 3 4 0 1 14
0.214 0.112 0.035 0.371 0.47 0.001* 0.441 0.056* 1* 1* 0.3* 0.050* 0.558* 0.672
192
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193
Table 4 Median laboratory values in different etiologies of hypercalcemia. Significance is by Mann–Whitney U test. p values are for comparisons between HHM and all other etiologies. MAHC, malignancy-associated hypercalcemia; HHM, humoral hypercalcemia of malignancy. Lab test (median)
Other etiology
MAHC
HHM
p
Total patients tested (percent)
Reference Interval
Ca, total (mg/dl) Ca, ionized (mg/dl) Albumin (g/dl) PTH (pg/mL) Mg (mg/dl) PO4 (mg/dl) Vit D 25-OH (ng/mL)
11.45 5.96 3.4 8 1.8 3.1 25
11.7 6.27 3.4 11 2 3 17
12.6 6.705 3 6 1.8 2.45 16
b0.001 b0.001 b0.001 b0.001 0.07 b0.001 0.009
463 (99) 355 (76) 383 (81) 455 (97) 375 (80) 393 (84) 305 (83)
8.6–10.3 4.5–5.1 3.6–5 14–72 1.4–2.5 2.3–4.3 30–100
detailed in Fig. 1. Each of the remaining 470 PTHrP tests which met inclusion criteria represented a unique patient. Evidence of malignancy was found for 242 of 470 patients (51%) with a PTHrP test ordered. 3.2. Etiology of malignancy-associated hypercalcemia Etiology of hypercalcemia was determined for all patients with PTHrP testing and evidence of malignancy (Table 1). After reviewing relevant laboratory data, pathology, radiology, and oncology notes, and discharge summaries, no etiology could be determined for 98 cases of MAHC (40%) and increased PTHrP contributed to 92 cases (38%) of MAHC. Overlapping etiologies were common with hypercalcemia attributed to multiple causes in 21% of cases. The most common overlapping etiologies were HHM and osteolytic bone lesions, cooccurring in 25 cases (10%). 3.3. Associations with humoral hypercalcemia of malignancy We investigated whether patient demographics or site of primary malignancy were significantly associated with HHM among patients with both hypercalcemia and evidence of malignancy. HMM was not associated with age, race or gender (Table 2). Consistent with previous studies [6,10] we found leukemia or lymphoma, cholangiocarcinoma, and cancers of bladder were most frequently associated with HHM (Table 3). Among all patients tested for PTHrP, values of calcium were significantly higher in patients with HHM (Table 4). Laboratory values of albumin, PTH, phosphate, and 25-OH vitamin D were all significantly lower in patients with HHM. Missing values were excluded on a per-test basis. Six of 91 patients with HHM had PTH ≥26 pg/ml, the cutoff suggested by Fritchie, et al. [8]. Of these, 4 had concurrent osteolytic bone lesions, one case was asymptomatic and untreated, and one patient expired during their diagnostic workup. The highest PTH value in any hypercalcemia caused exclusively by HHM was 21 pg/ml, which is consistent with Fritchie, et al. [8]. Increased PTHrP was present at initial malignancy diagnosis in 20% of cases. PTHrP was never increased outside of the context of malignancy. The diagnosis of HHM did not correlate with differences in length of stay or inpatient mortality (data not shown).
present study where patients were identified by PTHrP testing, previous studies identified patients with known malignancy and hypercalcemia. Although HHM is typically thought to manifest in advanced disease [12], 20% of our HHM patients had a new diagnosis of malignancy at the time of PTHrP testing. This could represent a significantly different patient population than previously studied [3]. We also found 21% of MAHC patients had more than one etiology for their hypercalcemia, and these patients were likely excluded from the previous analyses [4]. Finally, the difference in prevalence may reflect changes in ordering practices in intervening three decades between studies. This study supports a testing strategy of measuring ionized calcium and PTH prior to PTHrP testing [1,8]. Production of PTH is suppressed by increased serum PTHrP. The highest PTH of any patient with HHM as the sole etiology for their hypercalcemia was 21 ng/L. Twenty-four percent (128 of 524) of PTHrP assays were performed on patients with either normal or low serum calcium (n = 23) or PTH N21 ng/L (n = 105), both situations where PTHrP had no diagnostic utility. This study had several limitations. We did not find an etiology for 40% of MAHC cases. The majority of these were likely transient hypercalcemia, and none of these patients had any indication of follow-up for their hypercalcemia post-discharge. Nor did any of these patients have an increased PTHrP and thus would not affect HHM prevalence. This proportion of patients with an unknown or transient etiology is consistent with previous studies [3,13]. As we identified potential cases of HHM by PTHrP testing, our study population is dependent on the ordering practices at our institution. Although Barnes-Jewish Hospital is a large, tertiary care facility with a high volume of PTHrP testing, confirmation of these findings at other hospitals would be useful. While our overall sample size was large relative to previous studies, sample sizes for specific cancers were likely too small to find associations. In summary, the prevalence of HHM among patients with MAHC is lower than previously described. A larger sample size, inclusion of new diagnoses and concurrent etiologies, and changes in test ordering practices likely contributed to the difference in prevalence. Patients with hypercalcemia caused only by HHM had consistently low PTH. Calcium and PTH should be measured prior to PTHrP testing in cases where HHM is suspected.
4. Discussion References HHM is commonly thought to account for approximately 80% of MAHC [1,7]. However, a recent, smaller study identifying patients by PTHrP testing found a much lower prevalence of HHM, 32% among 47 patients with MAHC [8]. The present study similarly found that HHM only contributed to 38% of hypercalcemia among 242 cancer patients who underwent PTHrP testing at our institution. The commonly accepted prevalence of PTHrP is based on studies that either predate the identification of PTHrP [9] or studies validating the initial PTHrP immunoassay [10]. There are several possible reasons for the lower prevalence of HMM in the present study [1]. This study is substantially larger than previous investigations, identifying 242 patients with MAHC. Both of the previous investigations were smaller (50 [9] or 30 [10] patients). Unlike the
[1] A.F. Stewart, Clinical practice. Hypercalcemia associated with cancer, N. Engl. J. Med. 352 (2005) 373–379. [2] M.M. Shepard, J.W. Smith, Hypercalcemia, Am. J. Med. Sci. 334 (2007) 381–385. [3] F.W. Lafferty, Differential diagnosis of hypercalcemia, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 6 (Suppl. 2) (1991) S51–S59 (discussion S61). [4] G.J. Strewler, The physiology of parathyroid hormone-related protein, N. Engl. J. Med. 342 (2000) 177–185. [5] L.K. McCauley, T.J. Martin, Twenty-five years of PTHrP progress: from cancer hormone to multifunctional cytokine, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 27 (2012) 1231–1239. [6] J.E. Henderson, C. Shustik, R. Kremer, S.A. Rabbani, G.N. Hendy, D. Goltzman, Circulating concentrations of parathyroid hormone-like peptide in malignancy and in hyperparathyroidism, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 5 (1990) 105–113. [7] M.J. Horwitz, S.P. Hodak, A.F. Stewart, Asbmr. Chapter 67. Non-Parathyroid Hypercalcemia, Primer Metab Bone Dis Disord Miner Metab [Internet], John
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193 Wiley & Sons, Inc. 2008, pp. 307–312 ([cited 2015 Jul 22], Available from: http:// onlinelibrary.wiley.com/doi/10.1002/9780470623992.ch67/summary). [8] K. Fritchie, D. Zedek, D.G. Grenache, The clinical utility of parathyroid hormonerelated peptide in the assessment of hypercalcemia, Clin. Chim. Acta Int. J. Clin. Chem. 402 (2009) 146–149. [9] A.F. Stewart, R. Horst, L.J. Deftos, E.C. Cadman, R. Lang, A.E. Broadus, Biochemical evaluation of patients with cancer-associated hypercalcemia: evidence for humoral and nonhumoral groups, N. Engl. J. Med. 303 (1980) 1377–1383. [10] W.J. Burtis, T.G. Brady, J.J. Orloff, J.B. Ersbak, R.P. Warrell, B.R. Olson, et al., Immunochemical characterization of circulating parathyroid hormone-related protein in
193
patients with humoral hypercalcemia of cancer, N. Engl. J. Med. 322 (1990) 1106–1112. [11] A.A. Budayr, R.A. Nissenson, R.F. Klein, K.K. Pun, O.H. Clark, D. Diep, et al., Increased serum levels of a parathyroid hormone-like protein in malignancy-associated hypercalcemia, Ann. Intern. Med. 111 (1989) 807–812. [12] L. Santarpia, C.A. Koch, N.J. Sarlis, Hypercalcemia in cancer patients: pathobiology and management, Horm. Metab. Res. 42 (2010) 153–164 (Horm Stoffwechselforschung Horm Métabolisme). [13] D.M. Dent, J.L. Miller, L. Klaff, J. Barron, The incidence and causes of hypercalcaemia, Postgrad. Med. J. 63 (1987) 745–750.
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Incidence of humoral hypercalcemia of malignancy among hypercalcemic patients with cancer Jeffrey J. Szymanski 1, Zaher K. Otrock 1, Khushbu K. Patel, Mitchell G. Scott ⁎ Department of Pathology and Immunology, Washington University, St Louis, MO, United States
a r t i c l e
i n f o
Article history: Received 12 September 2015 Received in revised form 11 December 2015 Accepted 14 December 2015 Available online 17 December 2015 Keywords: Hypercalcemia Malignancy PTHrP Prevalence
a b s t r a c t Background: Malignancy-associated hypercalcemia (MAHC) is the most common cause of hypercalcemia among hospitalized patients. MAHC can result from the production of parathyroid hormone related peptide (PTHrP) which is known as humoral hypercalcemia of malignancy (HHM). HHM is commonly thought to account for approximately 80% of MAHC. Methods: We conducted a 12-year review of PTHrP testing at our institution to establish the prevalence of HHM among patients with MAHC. Results: A total of 524 PTHrP immunoassays were performed during the study period of which 470 tests qualified for inclusion in the analysis. Evidence of malignancy was found for 242 of 470 patients (51%). No etiology could be determined for 98 cases of MAHC (40%) and increased PTHrP contributed to 92 cases (38%) of MAHC. Age, race and gender were not associated with HHM. Increased PTHrP was observed at initial malignancy diagnosis in 20% of cases. PTHrP was never increased outside of the context of malignancy. Discussion: The prevalence of HHM among patients with MAHC is likely to be lower than previously described. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Malignancy-associated hypercalcemia (MAHC) occurs in 20–30% of cancer patients [1,2] and is the most common cause of hypercalcemia among hospitalized patients [3]. MAHC usually results from osteolytic bone lesions or from the production of parathyroid hormone related peptide (PTHrP), the latter of which is known as humoral hypercalcemia of malignancy (HHM). Nine of the 13 N-terminal amino acids of human PTHrP are homologous to those of parathyroid hormone, and PTHrP exerts its effects through binding to the parathyroid hormone 1 (PTH-1) receptor [4]. PTHrP is a peptide widely expressed throughout the body and circulates at low levels in the plasma of healthy individuals [5]. PTHrP is also ectopically secreted at much higher levels by some solid tumors and in some hematologic malignancies [6]. In these cancers, secreted PTHrP causes hypercalcemia via increased calcium absorption at the kidney and increased bone resorption [7]. Diagnostic workup of suspected HHM begins with the measurement of ionized calcium and PTH [1,8]. In patients with high ionized calcium and low PTH, measurement of PTHrP can confirm the diagnosis of HHM. HHM has been commonly thought to account for approximately 80% of MAHC [1,7]. This value is based on studies either predating the ⁎ Corresponding author at: 660 S. Euclid Ave, Campus Box 8118, St. Louis, MO 63110, United States. E-mail address: [email protected] (M.G. Scott). 1 JS and ZKO contributed equally to this manuscript.
http://dx.doi.org/10.1016/j.cca.2015.12.017 0009-8981/© 2015 Elsevier B.V. All rights reserved.
identification of PTHrP [9] or during validation of the original PTHrP immunoassay [10]. However, some early PTHrP immunoassay studies [11], and a recent, smaller study [8] reported lower prevalence of HHM. Our experience with PTHrP as a "resident-approval" test also challenged the accepted high prevalence of HHM in MAHC. 2. Methods 2.1. Study design To determine the prevalence of HHM in hospitalized patients with MAHC, we conducted a single-institution, cross sectional study of all PTHrP tests ordered at Barnes-Jewish Hospital in St. Louis over a twelve-year period from January 1, 2002 through December 31, 2013. PTHrP values were excluded from the study if they were not the first measurement for a given hospital admission, if they were performed on patients with normal or low calcium, or if associated patient records were unavailable. For each included PTHrP measurement, a chart review was performed to establish patient demographics, history of malignancy, relevant admitting laboratory values, etiology of hypercalcemia, and hospital admission outcome. This study was approved by the Washington University Institutional Review Board. 2.2. PTHrP immunoassays During the period of investigation, samples for PTHrP analysis were sent to one of four reference laboratories: Mayo Medical Laboratories
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193
191
Table 2 Demographics of patients with both elevated serum calcium and evidence of malignancy. Significance of age by Mann–Whitney U test. Significance of gender and race by Pearson chi-squared test. Age is in years. Gender and race values are total number of patients. HHM: humoral hypercalcemia of malignancy.
Fig. 1. Study design.
(311 assays), Quest Diagnostics (88 assays), Specialty Laboratories — Nichols Institute Diagnostics (30 assays), and ARUP Laboratories (41 assays). For all reference laboratories, PTHrP was determined using sandwich immunoassay. Results were evaluated against the reference interval provided by each laboratory and those exceeding the upper limit of the reference interval (1.9, 4.0, 4.7, or 1.3 pmol/l for the above mentioned reference laboratories, respectively) were considered increased. Positivity rates ranged from 7% to 20%. Specimens for PTHrP testing were collected according to the requirements given by each of the referral laboratories. 2.3. Clinical data collection PTHrP results and patient age and sex were collected from the hospital laboratory information system (Cerner Millennium). Patient race and admitting laboratory values were collected from the hospital electronic medical record (Clinical Desktop 2). Laboratory values included total and ionized calcium, albumin, PTH, phosphate, and 25-OH vitamin D. Total calcium (by colorimetric methods), total ionized calcium (by direct ion-selective electrode), albumin (by bromocresol green), PTH (by chemiluminescence enzyme immunoassay), phosphate (by ammonium molybdate), and 25-OH vitamin D (by immunoassay) were determined in-house. Length of stay and inpatient mortality were obtained from the hospital billing department. Type of malignancy and etiology of Table 1 Etiology of hypercalcemia among 242 patients with both elevated serum calcium and evidence of malignancy. Some patients had multiple causes of hypercalcemia. HHM: humoral hypercalcemia of malignancy. Etiology
Number of patients
Percent of patients
Undetermined HHM Osteolytic bone lesions Immobilization Primary hyperparathyroidism Tertiary hyperparathyroidism Dehydration Granulomatous disease Thiazide diuretics Calcium supplementation
98 92 66 8 6 2 2 2 1 1
40.5 38.0 27.3 3.3 2.5 0.8 0.8 0.8 0.4 0.4
HHM
Other etiology
p
Age (median)
59
60
0.528
Gender Male Female
55 36
93 58
0.859
Race Caucasian African American Other
59 31 1
98 51 2
0.96
hypercalcemia were determined by a manual review of laboratory values and patient charts by residents and fellows in the Department of Pathology and Immunology at Washington University. PTHrP was considered to contribute to MAHC if a patient had increased plasma total (N 10.3 mg/dl) or ionized (N5.1 mg/dl) calcium and an initial PTHrP value higher than the upper limit of the reference interval as provided by each of the reference laboratories. Osteolytic bone lesions were considered to contribute to MAHC if such lesions were documented on a radiographic imaging study or at autopsy. Other etiologies of hypercalcemia were identified through manual review of the electronic medical record. The etiology of a patient's hypercalcemia was considered unknown only if both PTH and PTHrP were normal and no cause was suggested or stated in pathology, radiology, or oncology notes or the discharge summary. 2.4. Data analysis All data analysis was performed in Microsoft Excel (ver 10, Microsoft) and SPSS Statistics (ver 22, IBM). Categorical variables were compared with Pearson chi-square or Fisher exact tests as appropriate. Association of individual laboratory values with etiology of hypercalcemia was assessed by Mann–Whitney U test. Laboratory results are presented as median values unless otherwise stated. 3. Results 3.1. PTHrP immunoassays A total of 524 PTHrP immunoassays were performed through Barnes-Jewish Hospital from January 1, 2002 to December 31, 2013 (Fig. 1). Fifty-four PTHrP tests (10%) did not meet inclusion criteria as Table 3 Site of primary malignancy among patients with both elevated serum calcium and evidence of malignancy. Three patients had multiple, concurrent cancer types. Significance by Fisher exact test or (*) for Pearson chi-squared test. HHM: humoral hypercalcemia of malignancy. Site of primary malignancy
Total
HHM
Other etiology
p
Head and neck Lung Leukemia or lymphoma Renal Breast Multiple myeloma Hepatocellular carcinoma Bladder Colon Pancreatic Thyroid Cholangiocarcinoma Ovarian Other
61 46 23 21 17 14 11 8 8 4 4 4 3 21
27 22 4 6 5 0 4 6 4 1 0 4 2 7
34 24 19 15 12 14 7 2 4 3 4 0 1 14
0.214 0.112 0.035 0.371 0.47 0.001* 0.441 0.056* 1* 1* 0.3* 0.050* 0.558* 0.672
192
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193
Table 4 Median laboratory values in different etiologies of hypercalcemia. Significance is by Mann–Whitney U test. p values are for comparisons between HHM and all other etiologies. MAHC, malignancy-associated hypercalcemia; HHM, humoral hypercalcemia of malignancy. Lab test (median)
Other etiology
MAHC
HHM
p
Total patients tested (percent)
Reference Interval
Ca, total (mg/dl) Ca, ionized (mg/dl) Albumin (g/dl) PTH (pg/mL) Mg (mg/dl) PO4 (mg/dl) Vit D 25-OH (ng/mL)
11.45 5.96 3.4 8 1.8 3.1 25
11.7 6.27 3.4 11 2 3 17
12.6 6.705 3 6 1.8 2.45 16
b0.001 b0.001 b0.001 b0.001 0.07 b0.001 0.009
463 (99) 355 (76) 383 (81) 455 (97) 375 (80) 393 (84) 305 (83)
8.6–10.3 4.5–5.1 3.6–5 14–72 1.4–2.5 2.3–4.3 30–100
detailed in Fig. 1. Each of the remaining 470 PTHrP tests which met inclusion criteria represented a unique patient. Evidence of malignancy was found for 242 of 470 patients (51%) with a PTHrP test ordered. 3.2. Etiology of malignancy-associated hypercalcemia Etiology of hypercalcemia was determined for all patients with PTHrP testing and evidence of malignancy (Table 1). After reviewing relevant laboratory data, pathology, radiology, and oncology notes, and discharge summaries, no etiology could be determined for 98 cases of MAHC (40%) and increased PTHrP contributed to 92 cases (38%) of MAHC. Overlapping etiologies were common with hypercalcemia attributed to multiple causes in 21% of cases. The most common overlapping etiologies were HHM and osteolytic bone lesions, cooccurring in 25 cases (10%). 3.3. Associations with humoral hypercalcemia of malignancy We investigated whether patient demographics or site of primary malignancy were significantly associated with HHM among patients with both hypercalcemia and evidence of malignancy. HMM was not associated with age, race or gender (Table 2). Consistent with previous studies [6,10] we found leukemia or lymphoma, cholangiocarcinoma, and cancers of bladder were most frequently associated with HHM (Table 3). Among all patients tested for PTHrP, values of calcium were significantly higher in patients with HHM (Table 4). Laboratory values of albumin, PTH, phosphate, and 25-OH vitamin D were all significantly lower in patients with HHM. Missing values were excluded on a per-test basis. Six of 91 patients with HHM had PTH ≥26 pg/ml, the cutoff suggested by Fritchie, et al. [8]. Of these, 4 had concurrent osteolytic bone lesions, one case was asymptomatic and untreated, and one patient expired during their diagnostic workup. The highest PTH value in any hypercalcemia caused exclusively by HHM was 21 pg/ml, which is consistent with Fritchie, et al. [8]. Increased PTHrP was present at initial malignancy diagnosis in 20% of cases. PTHrP was never increased outside of the context of malignancy. The diagnosis of HHM did not correlate with differences in length of stay or inpatient mortality (data not shown).
present study where patients were identified by PTHrP testing, previous studies identified patients with known malignancy and hypercalcemia. Although HHM is typically thought to manifest in advanced disease [12], 20% of our HHM patients had a new diagnosis of malignancy at the time of PTHrP testing. This could represent a significantly different patient population than previously studied [3]. We also found 21% of MAHC patients had more than one etiology for their hypercalcemia, and these patients were likely excluded from the previous analyses [4]. Finally, the difference in prevalence may reflect changes in ordering practices in intervening three decades between studies. This study supports a testing strategy of measuring ionized calcium and PTH prior to PTHrP testing [1,8]. Production of PTH is suppressed by increased serum PTHrP. The highest PTH of any patient with HHM as the sole etiology for their hypercalcemia was 21 ng/L. Twenty-four percent (128 of 524) of PTHrP assays were performed on patients with either normal or low serum calcium (n = 23) or PTH N21 ng/L (n = 105), both situations where PTHrP had no diagnostic utility. This study had several limitations. We did not find an etiology for 40% of MAHC cases. The majority of these were likely transient hypercalcemia, and none of these patients had any indication of follow-up for their hypercalcemia post-discharge. Nor did any of these patients have an increased PTHrP and thus would not affect HHM prevalence. This proportion of patients with an unknown or transient etiology is consistent with previous studies [3,13]. As we identified potential cases of HHM by PTHrP testing, our study population is dependent on the ordering practices at our institution. Although Barnes-Jewish Hospital is a large, tertiary care facility with a high volume of PTHrP testing, confirmation of these findings at other hospitals would be useful. While our overall sample size was large relative to previous studies, sample sizes for specific cancers were likely too small to find associations. In summary, the prevalence of HHM among patients with MAHC is lower than previously described. A larger sample size, inclusion of new diagnoses and concurrent etiologies, and changes in test ordering practices likely contributed to the difference in prevalence. Patients with hypercalcemia caused only by HHM had consistently low PTH. Calcium and PTH should be measured prior to PTHrP testing in cases where HHM is suspected.
4. Discussion References HHM is commonly thought to account for approximately 80% of MAHC [1,7]. However, a recent, smaller study identifying patients by PTHrP testing found a much lower prevalence of HHM, 32% among 47 patients with MAHC [8]. The present study similarly found that HHM only contributed to 38% of hypercalcemia among 242 cancer patients who underwent PTHrP testing at our institution. The commonly accepted prevalence of PTHrP is based on studies that either predate the identification of PTHrP [9] or studies validating the initial PTHrP immunoassay [10]. There are several possible reasons for the lower prevalence of HMM in the present study [1]. This study is substantially larger than previous investigations, identifying 242 patients with MAHC. Both of the previous investigations were smaller (50 [9] or 30 [10] patients). Unlike the
[1] A.F. Stewart, Clinical practice. Hypercalcemia associated with cancer, N. Engl. J. Med. 352 (2005) 373–379. [2] M.M. Shepard, J.W. Smith, Hypercalcemia, Am. J. Med. Sci. 334 (2007) 381–385. [3] F.W. Lafferty, Differential diagnosis of hypercalcemia, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 6 (Suppl. 2) (1991) S51–S59 (discussion S61). [4] G.J. Strewler, The physiology of parathyroid hormone-related protein, N. Engl. J. Med. 342 (2000) 177–185. [5] L.K. McCauley, T.J. Martin, Twenty-five years of PTHrP progress: from cancer hormone to multifunctional cytokine, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 27 (2012) 1231–1239. [6] J.E. Henderson, C. Shustik, R. Kremer, S.A. Rabbani, G.N. Hendy, D. Goltzman, Circulating concentrations of parathyroid hormone-like peptide in malignancy and in hyperparathyroidism, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 5 (1990) 105–113. [7] M.J. Horwitz, S.P. Hodak, A.F. Stewart, Asbmr. Chapter 67. Non-Parathyroid Hypercalcemia, Primer Metab Bone Dis Disord Miner Metab [Internet], John
J.J. Szymanski et al. / Clinica Chimica Acta 453 (2016) 190–193 Wiley & Sons, Inc. 2008, pp. 307–312 ([cited 2015 Jul 22], Available from: http:// onlinelibrary.wiley.com/doi/10.1002/9780470623992.ch67/summary). [8] K. Fritchie, D. Zedek, D.G. Grenache, The clinical utility of parathyroid hormonerelated peptide in the assessment of hypercalcemia, Clin. Chim. Acta Int. J. Clin. Chem. 402 (2009) 146–149. [9] A.F. Stewart, R. Horst, L.J. Deftos, E.C. Cadman, R. Lang, A.E. Broadus, Biochemical evaluation of patients with cancer-associated hypercalcemia: evidence for humoral and nonhumoral groups, N. Engl. J. Med. 303 (1980) 1377–1383. [10] W.J. Burtis, T.G. Brady, J.J. Orloff, J.B. Ersbak, R.P. Warrell, B.R. Olson, et al., Immunochemical characterization of circulating parathyroid hormone-related protein in
193
patients with humoral hypercalcemia of cancer, N. Engl. J. Med. 322 (1990) 1106–1112. [11] A.A. Budayr, R.A. Nissenson, R.F. Klein, K.K. Pun, O.H. Clark, D. Diep, et al., Increased serum levels of a parathyroid hormone-like protein in malignancy-associated hypercalcemia, Ann. Intern. Med. 111 (1989) 807–812. [12] L. Santarpia, C.A. Koch, N.J. Sarlis, Hypercalcemia in cancer patients: pathobiology and management, Horm. Metab. Res. 42 (2010) 153–164 (Horm Stoffwechselforschung Horm Métabolisme). [13] D.M. Dent, J.L. Miller, L. Klaff, J. Barron, The incidence and causes of hypercalcaemia, Postgrad. Med. J. 63 (1987) 745–750.