Familial isolated hyperparathyroidism: clinical and genetic characteristics of 36 kindreds.

Abstract

Introduction Abbreviations used in this paper: ANOVA, analysis of variance, CaSR, calcium-sensing receptor, CASR, calcium-sensing receptor gene, CT, computed tomography, FIH, familial isolated hyperparathyroidism, FHH, familial hypocalciuric hypercalcemia, HPT, hyperparathyroidism, HPT-JT, hyperparathyroidism-jaw tumor syndrome, MD-CRI, Molecular Diagnostics program of the Children’s Research Institute (Washington, DC), MEN1, multiple endocrine neoplasia type 1, MEN2A, multiple endocrine neoplasia type 2A, MIM, Mendelian Inheritance in Man, NIDDK, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, National Institutes of Health, NS, not significant, PTX, parathyroidectomy, SEM, standard error of the mean Familial hyperparathyroidism (HPT) encompasses a clinically and genetically heterogeneous group of disorders. Syndromes with familial HPT include multiple endocrine neoplasia type 1 (MEN1) (Mendelian Inheritance in Man [MIM] 131100 1) (63,87), multiple endocrine neoplasia type 2A (MEN2A) (MIM 171400) (42,79,86), familial hypocalciuric hypercalcemia (FHH) (MIM 145980, 145981, 600740) also known as familial benign hypercalcemia (38,64), and the hyperparathyroidism-jaw tumor syndrome (HPT-JT; HRPT2) (MIM 145001) (45). Familial isolated hyperparathyroidism 2 (FIH; HRPT1) (MIM 145000) is a subgroup of familial HPT that can result from the incomplete expression of a syndromic form of familial HPT or from full expression of other entities (Figure 1). It is unknown how many as yet unrecognized clinical entities, including mutant genotypes, can also present as FIH.Fig. 1: The relationship as a Venn diagram among familial forms of hyperparathyroidism that may present as familial isolated hyperparathyroidism (FIH). The dashed circle represents the set of patients presenting with the provisional diagnosis of FIH (see text footnote 2) and encompasses the 36 families enrolled in this study. Contained entirely within this dashed circle is a subset of families who have subsequently been thoroughly evaluated for, but lack findings diagnostic of, multiple endocrine neoplasia type 1 (MEN1), familial hypocalciuric hypercalcemia (FHH), and hyperparathyroidism-jaw tumor syndrome (HPT-JT) (FIH; in a solid circle). Subsets of patients with incomplete expression of MEN1, FHH, and HPT-JT (the total set of patients in each syndrome represented by a solid circle) can also present with the FIH phenotype. The distinction between the FIH category and the syndromic categories arbitrarily depends on the sensitivity of diagnostic tests used to detect the syndrome. Multiple endocrine neoplasia type 2A (MEN2A) is a familial form of hyperparathyroidism seldom if ever presenting as FIH. See text for additional details and references. Within each circle representing a defined syndrome are included the genetic locus (or loci in the case of FHH) of the syndromic trait and the responsible gene product (if known). An asterisk next to the genetic locus indicates that the gene and gene product are unknown for the form of familial hyperparathyroidism mapping to this site. The relationship among the patient groups is intended to be qualitative, and the area of each circle and the area of overlap between circles are not intended to be proportional to their encountered or predicted values.MEN1 is an autosomal dominant disorder characterized by endocrine and nonendocrine tumors, most strikingly involving the parathyroids, enteropancreatic endocrine system, and pituitary. Because FIH is seen less frequently than full expressions of MEN1, because HPT is the earliest and most frequent endocrinopathy in MEN1, and because even some large families with an apparent phenotype of FIH ultimately express MEN1, we (59,67) and others (2,61) previously speculated that most kindreds with FIH were occult expressions of MEN1. The gene responsible for MEN1 has been cloned (15), leading to powerful gene sequencing methods applicable to MEN1, FIH, and other conditions (63). MEN2A, unlike MEN1, is not typically a consideration in the differential diagnosis of FIH, because the higher penetrance of medullary thyroid carcinoma and pheochromocytoma than of HPT in MEN2A dominates the clinical presentation in a family (42,79,86). FHH is an autosomal dominant trait usually causing mild HPT (62) with relative hypocalciuria; hypercalcemia in FHH is highly penetrant at all ages, even in the perinatal period (64). Mild hypermagnesemia is sometimes seen in FHH but is unusual in other forms of primary HPT (53,64). FHH cases almost always remain hypercalcemic following standard subtotal parathyroidectomy (PTX) (64). FHH always presents initially as FIH, but the syndromic diagnosis of FHH is frequently straightforward, especially in kindreds with many affected members (54,64). Most cases of FHH result from a loss-of-function mutation in the gene encoding the calcium-sensing receptor (CaSR 3) (MIM 601199) on the long arm of chromosome 3 (10,38,46,77). However 2 undiscovered genes have been implicated in rare kindreds with FHH; 1 gene at chromosome 19p (HHC2, MIM 145981) (34) and 1 gene at 19q (HHC3, MIM 600740) (56). CASR testing for germline mutation has not been used widely, having been restricted to several research studies (reviewed in references 10,38). HPT-JT syndrome is an autosomal dominant disorder with high but incomplete penetrance of HPT. Its commonest features are parathyroid adenoma, parathyroid carcinoma, and/or fibroosseous jaw tumors (45). Renal cysts (98,99) and solid renal tumors (32,47,98) have also been associated with HPT-JT. HPT-JT with parathyroid tumors that are (60) or are not (21,45,50,106) cystic and FIH associated with parathyroid carcinoma without evident jaw tumors (99,108,110) are kindred features likely to share a common genetic basis. The trait in the few large kindreds reported with such disorders has been linked to 1q25–q31 (95,98,99,107,110), but the gene responsible for HPT-JT has not yet been identified. Based on the few families reported with HPT-JT, HPT-JT seems far less prevalent than MEN1, MEN2A, or FHH. The few previous studies of kindreds with FIH did not evaluate for HPT-JT and did not exclude occult MEN1 (41) and occult FHH (41,100) with full clinical and biochemical evaluations. Furthermore, there has been no large series of FIH from a single institution. We present here a detailed analysis of 36 kindreds with a provisional diagnosis of FIH from 1 medical center. Patients and Methods Patients The provisional diagnosis of FIH and entry into this study were based upon biochemical evidence of primary HPT in the proband plus 1 or more first-degree relatives; this also required at least 1 family member with histopathologically verified abnormal parathyroid tissue. Probands were evaluated through the Metabolic Diseases Branch of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) at the Clinical Center of the National Institutes of Health (NIH), Bethesda, MD, between 1971 and 2000. Most probands (20 of 36) were referred for evaluation and treatment of persistent or recurrent HPT following prior PTX. (Recurrent HPT was defined as that which followed a documented interval of 3 months or more of normocalcemia after an earlier PTX.) Relatives of a proband were tested either at NIH or at local facilities. Three probands were referred mainly for MEN1 gene germline mutational analysis and evaluated without any family member visiting the NIH. All patient testing was performed with informed consent under protocols approved by the NIDDK Institutional Review Board. Kindreds were excluded if the diagnosis of MEN1, FHH, or HPT-JT had been proven or considered likely before proband referral to NIH or before the design of this clinical analysis in 1998. Approximately 100 kindreds with MEN1 (of which few if any had the FIH phenotype) were diagnosed by biochemical and/or mutational criteria and evaluated through the Metabolic Diseases Branch between 1971 and 2000, and thus excluded, as were approximately 60 kindreds with FHH. Two kindreds with the prior established diagnosis of HPT-JT were also excluded. We could not locate living affected members or gather adequate data from deceased affected members from 3 possible FIH kindreds whose probands had been evaluated at the NIH between 1974–1977 (66), so these families were not included. The index case was generally the proband. If the proband was deceased or otherwise unavailable, another affected individual from the same kindred was designated as the index case, but those data were not treated as proband data. Wherever possible clinical data were also obtained on additional affected members from each kindred and used in the consideration of a syndromic diagnosis. Portions of kindreds 2862 (93), 24,700 (59), and 28,200 (88) were previously reported. The absence of germline MEN1 mutation in 5 FIH probands (from kindreds 4318, 5780, 6325, 7751, and 9462) was previously reported (1). Surgical specimens Primary parathyroid gland tumors were considered for this study to include cervical or mediastinal tumors found at any operation, and excluded local or distant metastatic tumors. Parathyroid tumor surgical specimens from NIH and, when possible, from outside institutions were reviewed by the Surgical Pathology Department of the National Cancer Institute. Members of the Department of Otolaryngologic and Endocrine Pathology of the Armed Forces Institute of Pathology (Washington, DC) confirmed parathyroid cancer in the proband of kindred 30,300, primary carcinomas in 2 parathyroid glands of the uncle of the proband in kindred 417 (II-3 in Figure 2), and primary carcinoid tumor of the gall bladder in the proband of kindred 9462.Fig. 2: Three kindreds with the hyperparathyroidism-jaw tumor syndrome. Individuals II-1 and II-2 in kindred 417 are fraternal twins. The pedigree of kindred 2862 has been updated from an earlier version (93). Square symbols indicate males, and round symbols indicate females. A diagonal slash mark through the symbol means the individual is deceased. The arrow indicates the proband in each kindred. The hatched markings in the upper left quadrant of individual II-1 in kindred 2862 indicate suspected severe hyperparathyroidism (based on the presence of a giant cell tumor of the tibia and renal calcifications discovered at autopsy [93]). Blank symbols represent family members not evaluated for this study. The black dot at the bottom of an otherwise blank symbol indicates that the serum calcium was normal at the latest screening in an individual never known to have hypercalcemia.Laboratory data Testing at NIH was through the Clinical Chemistry Service of the Department of Laboratory Medicine and also through the Diagnostic Radiology Department of the NIH Clinical Center. Testing of calcium and creatinine in serum and in urine from 24-hour collections employed standard automated techniques. Relative hypocalciuria was diagnosed if the ratio of the renal clearance of calcium to that of creatinine was less than 0.01 (64). Hypermagnesemia was diagnosed if an elevated serum magnesium level was confirmed on repeat testing after the magnesium level obtained during the initial evaluation of hypercalcemia was found to be 0.99 mmol/L or above (normal, 0.75–1.00 mmol/L). Plasma intact PTH and serum gastrin and serum prolactin measurements were by immunoassay. Screening for jaw tumors employed orthopantographic X-rays and/or computed tomography (CT) scan of the mandible and maxilla. Evaluation of the kidneys employed standard ultrasound or, in a few cases, CT scan with contrast. When multiple blood and urine chemistry data were available for a patient who had undergone repeated PTX, the earliest available preoperative data were used. Otherwise data from the initial NIH evaluation were used. Germline mutation analysis of the MEN1 gene DNA for analysis was extracted from anticoagulated whole blood. Mutational analysis of the MEN1 gene was performed in 1 of 2 ways. The first method, performed in NIDDK, employed PCR amplification of segments encompassing the entire open reading frame and intron-exon boundaries, screening of sequence by dideoxy fingerprinting, followed by DNA sequencing limited to abnormal segments (15), with minor alterations available at http://www.niddk.nih.gov/welcome/releases/primers.htm. The second method was through the Molecular Diagnostics program of the Children’s Research Institute (MD-CRI) (affiliated with the Children’s National Medical Center, Washington, DC) (111); this method employed PCR amplification and sequencing of segments encompassing the entire open reading frame and intron-exon boundaries of the MEN1 gene. To compare the 2 assays for mutations in the MEN1 gene, 15 samples from MEN1 probands, including 12 with different MEN1 gene mutations and 3 without mutation known from prior NIDDK analysis, were provided to MD-CRI as unknowns. All paired results of the 2 assays were the same. Germline mutation testing of the calcium-sensing receptor gene Mutational analysis was performed on DNA isolated from anticoagulated whole blood. Exons 2–7 encompassing the entire coding region of the CASR gene along with their intron-exon boundaries on chromosome 3 were amplified individually, followed by direct sequencing (19). Missense mutations in DNA from kindreds 2862, 5780, 10,147, 23,300, and 28,300 were confirmed by restriction enzyme digestion and were not present in 100 chromosomes in a DNA panel of 50 unrelated normal individuals. The deletion in exon 4 of kindred 1214 was confirmed by sequencing of subcloned DNA. Diagnostic classification of the kindreds The kindreds provisionally diagnosed with FIH were subgrouped after completion of the biochemical, gene mutational, and imaging studies according to the following diagnostic criteria (see text for references to each syndrome). MEN1 or occult MEN1 would have been diagnosed if a member with HPT also had proven or strongly suspected gastrinoma or prolactinoma and/or a sequence change in the MEN1 gene likely to cause loss of menin function (87). Diagnosis of CASR-mutation-positive was made if the index case with HPT had a sequence change in the CASR gene (see above) known or likely to cause loss of CaSR function. Whenever possible close relatives of hypercalcemic patients with CASR sequence changes were also screened biochemically and by gene analysis to look for linkage of the trait to the altered genotype. A category of MEN1 without MEN1 mutation or a category of FHH without CASR mutation was theoretically recognizable, but no family had features that justified populating either category. The diagnosis of HPT-JT in a family provisionally diagnosed with FIH required 1 individual with 2 of the following, or each of at least 2 individuals with a different 1 of the following manifestations: parathyroid carcinoma, fibroosseous jaw tumors, and bilateral renal cysts (2 or more cysts per kidney). Kindreds that did not meet the criteria for 1 of the 3 syndromic categories above remained classified as FIH. Statistical analysis Statistical analysis utilized GraphPad InStat software (version 3.0a for Macintosh, Graphpad Software, San Diego, CA). For comparison of continuous variables between 2 patient subgroups, the unpaired t test was used (nonparametric analysis). The Fisher exact test (a contingency table) was used to compare a categorical variable between 2 patient subgroups. Multiple means analysis, comparing a continuous variable among the 3 FIH patient subgroups, utilized 1-way analysis of variance (ANOVA) (nonparametric analysis). Post-ANOVA comparison between 2 of the 3 patient subgroups employed the Tukey-Kramer multiple comparisons test. Comparison of categorical variables among the 3 patient subgroups employed the chi-square test for independence with 2 degrees of freedom. Variation among individual values was expressed as the standard error of the mean (SEM). Results Overview of probands, kindreds, and nonparathyroid tumors Thirty-six apparently unrelated kindreds, encountered over 30 years, received a provisional diagnosis of FIH and underwent further testing for this study (see Figure 1). Among 170 living first-degree relatives of the probands, 104 (61%) were screened by serum calcium determination and sometimes additional tests. The clinical characteristics of each proband and an overview of each kindred are presented (Table 1). The families are grouped into 3 categories according to the ultimate results of clinical and genetic tests: HPT-JT, CASR-mutation-positive, or FIH. No family was given the diagnosis of MEN1 (see Patients and Methods, and below). The median number of affected individuals per kindred is only 2 (mean, 3.0; range, 2–11). Among 106 affected members in the 36 kindreds with a provisional diagnosis of FIH the gender ratio was 1.5:1 female:male (see Table 1), a lower proportion of females than among sporadic cases of HPT (2.7:1, n = 253; p < 0.05) (57). The majority of probands (56%) had been asymptomatic at the time of initial presentation outside NIH, with hypercalcemia usually an incidental finding on multichannel chemistry screens. This pattern of presentation is similar to that for primary hyperparathyroidism in recent decades (33).TABLE 1: Familial isolated hyperparathyroidism: Overview of probands and other affected members in provisionally diagnosed kindredsTABLE 1: —ContinuedTABLE 1: —ContinuedThe average age at diagnosis of HPT among the 36 probands (39 ± 3 [mean ± SEM] yr) was at least a decade younger than that for sporadic HPT patients treated between 1980 and 1997 (53 ± 1 yr, n = 253; p < 0.0001) (57), but not significantly different from that of nonprospectively identified patients in MEN1 kindreds (38 ± 3 yr, n = 22) (92), probands from typical FHH kindreds (48 ± 6.1 yr, n = 6) (68), and probands from HPT-JT kindreds (33 ± 3 yr, n = 18) (14,21,26,29,43,44,47,50,60,82,98,99,102,106,108,113). The average ages at diagnosis of HPT among probands from the HPT-JT subgroup (28 ± 11 yr, n = 3), the CASR-mutation-positive subgroup (37 ± 7 yr, n = 5), and the FIH subgroup (40 ± 3 yr, n = 28) were similar by 1-way ANOVA (F = 0.8) (see Table 1). Sixteen of the 36 kindreds with a provisional diagnosis of FIH contained affected members with benign or malignant tumors of nonparathyroid tissue (see Table 1). Tumors in affected members from 2 or more kindreds included breast cancer (n = 8 cases among 6 kindreds), colorectal cancer (n = 5 cases among 4 kindreds), jaw tumors (n = 5 cases among 3 kindreds) (Figure 2), lipoma (n = 4 cases in 3 kindreds), renal angiomyolipoma (n = 2 cases in 2 kindreds), nonmedullary thyroid cancer (n = 2 cases in 2 kindreds), and malignant dermal melanoma (72) (n = 2 cases in 2 kindreds). (Besides the 4 patients from 3 kindreds in the FIH subgroup with both HPT and lipomas, 2 additional members of kindred 8715 had lipomas but not HPT [Figure 3]. Since these 2 members did not have HPT they are considered unaffected for the purposes of this study.) In 1 kindred from the FIH subgroup (8715) parathyroid tumors and breast cancer cosegregated in females over 3 generations (see Figure 3), and in another kindred with HPT-JT there were coincident parathyroid tumors and colorectal cancer in 2 members (kindred 417, see Table 1).Fig. 3: Kindred 8715 with cosegregating breast cancer and hyperparathyroidism in 3 generations. Only individuals I-2, II-3, II-4, and III-11 are scored as affected for the purposes of this study (see Results section). See the legend for Figure 2 for explanation of symbols and markings.Testing directed at MEN1 The results of biochemical tests to explore for a diagnosis of MEN1 in members of each of the 36 kindreds are summarized (Table 2). Prior diagnosis of gastrinoma or prolactinoma in a member with HPT was an exclusion criterion for that family from this study. However, neither the 36 index cases nor any of their known affected relatives in the families provisionally diagnosed with FIH had these tumors recognized on follow-up or after the more detailed study testing (63). Among 31 kindreds 50 affected individuals had fasting gastrin levels tested and only 3 showed mild elevations (range, 244–721 pg/mL; normal, 0–100 pg/mL) (see Table 2). One of these patients underwent endoscopy and gastric acid analysis; no evidence of peptic ulcer disease or excessive gastric acid secretion was found thus excluding the Zollinger-Ellison syndrome. The other 2 patients with hypergastrinemia had no signs or symptoms of peptic ulcer disease, but they declined gastric acid analysis, although it was recommended. Among 50 affected individuals tested for serum prolactin level from 31 kindreds, only 2 showed values that were elevated (35 and 96 μg/L, respectively; normal, 1–11 μg/L) (see Table 2), although not into the higher range typically associated with prolactinoma (>200 μg/L ) (80). At the time of testing, 1 (proband of kindred 10,147) was taking neuroleptics and had a normal pituitary magnetic resonance imaging (MRI). Pituitary MRI on the other hyperprolactinemic patient (niece of proband of kindred 2862) revealed a 2-mm pituitary cyst and no other evidence of pituitary adenoma. Furthermore these 2 cases were in 2 different families found to have an HPT syndrome other than MEN1 (see Table 2).TABLE 2: Biochemical, DNA mutational, and imaging test results in affected members of kindreds with the provisional diagnosis of familial isolated hyperparathyroidismTABLE 2: —ContinuedTABLE 2: —ContinuedNone of the 36 index cases showed germline MEN1 gene mutation. Several common benign polymorphisms in the MEN1 gene were found including D418D (n = 17), A541T (n = 3), R171Q (n = 3), and S145S (n = 1) (1). Each polymorphism was heterozygous with the wildtype allele (data not shown). Testing directed at FHH None of the kindreds carried a diagnosis of FHH at the time of enrollment into this study. Despite this exclusion, clinical testing was performed to try to identify kindreds that might have unrecognized or atypical FHH (64). Calcium levels with or without creatinine urinary excretion were determined in 50 affected and hypercalcemic individuals in 33 kindreds, and all pedigrees were reviewed and extended. This included further testing, where possible, for evidence of hypercalcemia in members including those under the age of 10 years. One or more hypercalcemic individuals from each of 6 kindreds in this study demonstrated relative hypocalciuria; 3 of these kindreds were classified as CASR-mutation-positive. Unexpectedly, probands from 3 of the 5 FIH kindreds diagnosed as CASR-mutation-positive (see below) were hypercalciuric (see Table 2). Four families including 2 CASR-mutation-positive kindreds had hypercalcemic members less than 10 years old, the youngest being 4 months of age (kindred 5780) (see Table 2 and Figure 4).Fig. 4: Kindred 5780 with a diagnosis of familial hypocalciuric hypercalcemia associated with an arginine886 to proline missense mutation in the calcium-sensing receptor. See the legend for Figure 2 for explanation of symbols and markings.Serum magnesium levels were determined on 52 hypercalcemic patients from 32 of the 36 kindreds in this study (see Table 2). Individuals from 4 kindreds demonstrated mild hypermagnesemia (53,64), including 3 kindreds subsequently categorized as CASR-mutation-positive (see below). Significant differences were observed among diagnostic subgroups of the 36 study kindreds when the mean serum magnesium levels of tested hypercalcemic members were compared: CASR-mutation-positive, 1.02 ± 0.04 mmol/L, n = 10; HPT-JT, 0.84 ± 0.04 mmol/L, n = 8; and FIH, 0.84 ± 0.02 mmol/L, n = 34 (p < 0.0001, F = 12.2; Tukey-Kramer comparisons CASR-mutation-positive versus HPT-JT p < 0.01, CASR-mutation-positive versus FIH p < 0.001, HPT-JT versus FIH p = NS) (see Table 2). Only 1 kindred had overall clear features of FHH. Although its proband was hypercalciuric, kindred 5780 revealed relative hypocalciuria in other affected members (n = 4), hypercalcemic members less than 10 years old (n = 4), mild hypermagnesemia in hypercalcemic members (n = 4), and persistent hypercalcemia following subtotal parathyroidectomy (n = 2) (see Tables 2 and 3, Figure 4).TABLE 3: Parathyroid surgeries and operative outcome in affected members of kindreds with the provisional diagnosis of familial isolated hyperparathyroidismTABLE 3: —ContinuedTABLE 3: —ContinuedGermline mutational analysis was performed on the CASR gene for an index case from each of the 36 kindreds. DNA changes were found that resulted in 5 missense amino acid substitutions (T445A, R886P, R220W, L159P, E250K) and 1 11 bp deletion in exon 4 resulting in a frameshift (V268del-11X273) (see Table 2). Other common and benign CASR polymorphisms (35,83) were found among the 36 index cases in this study, including A986S (n = 15), G990R (n = 3), and Q1011E (n = 2). Each mutation and benign polymorphism was heterozygous with the wildtype allele (data not shown). Five of the kindreds in this study were classified as CASR-mutation-positive: 4 kindreds with nonconservative CASR missense mutations and 1 kindred with an 11 bp frameshift deletion in exon 4 of the CASR gene. In kindred 5780, all 7 tested hypercalcemic members carried the R886P mutation (see Table 2 and Figure 4) supporting its causative role in the syndrome; further, 3 normocalcemic first-degree relatives had no CASR mutation (data not shown). In a 6th kindred (2862) the CASR change, encoding a conservative amino acid substitution, was classified as a rare benign polymorphism; that kindred was diagnosed with HPT-JT based on additional testing results (see Table 2 and below). Imaging tests directed at HPT-JT Jaw radiographs were obtained on 38 affected individuals from 26 of the kindreds (see Table 2). Five patients with jaw lesions were identified, including 4 cases from 2 HPT-JT kindreds and 1 FIH patient. Two of these patients (1 HPT-JT and 1 FIH) had lesions that were occult and recognized only during screening for this study, while 3 others had histories of jaw surgery for tumors, information not originally judged relevant to HPT or included in the referring medical record. In kindred 2862, 3 sibs were found to have jaw lesions, in each case determined to be cementoossifying fibrous tumor by histopathologic analysis (see Tables 1, 2;Figure 2); a previous evaluation of this kindred included jaw imaging of the proband only, with normal findings (93). The father of the proband in kindred 27,000 had multiple mandibular and maxillary surgeries to remove jaw “cysts” in the 1970s; medical records or pathology specimens documenting the histopathology could not be retrieved (see Tables 1, 2;Figure 2). One individual from kindred 21,300 had a small mixed radiolucent/radiopaque mandibular lesion with fibroosseous histology on biopsy. Kidney imaging data were obtained from 29 affected patients representing 22 kindreds (see Table 2). Bilateral renal cysts were found in 6 members of 3 kindreds, each with HPT-JT: 3 individuals from kindred 417, 2 individuals from kindred 2862, and the proband from kindred 27,000 (see Table 2;Figure 2). In the latter case the renal cysts were previously unrecognized. No bilateral renal cysts were found in patients from the CASR-mutation-positive or FIH subgroups. No renal hamartoma, nephroblastoma, or renal cell carcinoma—lesions also reported in occasional association with HPT-JT (32,47,98) —was identified. The probands from 2 FIH kindreds had renal angiomyolipoma diagnosed by ultrasound and/or CT scan (12) (see Table 2). Such renal tumors are frequently associated with tuberous sclerosis (73) and have been rare concomitants of MEN1 (22,91). In 3 of the 36 study kindreds (417, 2862, 27,000) the coexpression of bilateral renal cysts with parathyroid cancer and/or jaw tumors established the diagnosis of HPT-JT (see Table 2;Figure 2). Several other kindreds classified as FIH had weaker evidence for the diagnosis of HPT-JT, such as kindred 21,300 discussed above, kindreds 24,700 (59) and 30,300 in which the proband had parathyroid cancer (see Tables 1 and 3), and kindred 28,200 (88) with cystic parathyroid tumors in 3 members (60) (see below). Findings and outcomes from parathyroid surgery Parathyroid operative reports, histopathologic results, and/or postoperative medical records for 76 of 79 operated patients were reviewed (Table 3) and summarized for each diagnostic subgroup (Table 4). The number of operations per case was highest in the HPT-JT subgroup because of the high prevalence of parathyroid cancer in HPT-JT (see Table 4 and below). Nearly half of the operated HPT-JT patients had parathyroid cancer, and 1 HPT-JT patient (kindred 417, II-3 in Figure 2) had primary parathyroid carcinoma in 2 separate glands at initial cervical exploration.TABLE 4: Parathyroid surgical findings and late outcome among operated patients from kindreds with the provisional diagnosis of familial isolated hyperparathyroidismQuantitative information about primary parathyroid tumor size was available in 69 of the 76 operated patients from the 36 study kindreds provisionally diagnosed as FIH (see Tables 3 and 4). The size of the largest primary tumor was greater in the HPT-JT than in the CASR-mutation-positive subgroup (see Table 4). Omission of parathyroid cancer from the HPT-JT and FIH subgroups eliminates the statistical significance of the size differences in maximum dimension of primary parathyroid tumor among the 3 subgroups (see Table 4). This effect on significance but not on average value was due mainly to the reduction in the number of data points when parathyroid cancers were excluded from the HPT-JT group. There was no statistical difference in primary tumor size between parathyroid cancer and benign parathyroid tumors within the HPT-JT (3.3 ± 0.4 cm, cancer, n = 5; 3.1 ± 0.4 cm, benign, n = 6) and FIH (2.3 ± 0.8 cm, cancer, n = 2; 2.3 ± 0.2 cm, benign