Isolated Familial Somatotropinomas: Does the Disease Map to 11q13 or to 2p16?

Constantine A. Stratakis and Lawrence S. Kirschner

Unit on Genetics and Endocrinology (UGEN) Developmental Endocrinology Branch National Institute of Child Health and Human Development National Institutes of Health Bethesda, Maryland 20892-1862

To the editor:

We read with great interest the article by Gadelha et al. (1) in the February issue of JCEM. The authors had previously described two families with apparent isolated familial somatotropinomas (IFS) and reported loss of heterozygosity (LOH) of their tumor tissue at 11q13 (2). They then examined their families for possible linkage to the multiple endocrine neoplasia type 1 (MEN 1) locus at 11q13 and the Carney complex (CNC) loci at 2p16 and 17q22-24. To perform this analysis, Gadelha et al. (1) used a program by Rohde et al. (3), which adjusts logarithm of odds (LOD) scores for LOH data. The modified maximum LOD score determined in this way was 4.0 for 11q13 (family B). Although it is likely that another gene with tumor suppression function in endocrine tissues is located on 11q13 (4), several questions are raised regarding the actual linkage of the families reported by Gadelha et al. (1) to 11q13 vs. 2p16.

We have recalculated the LOD scores for 11q and 2p using the germ-line alleles reported by Gadelha et al. (1); this analysis yielded two-point LOD scores that are strongly positive for both loci, but not conclusive (Table 1Go). We did not include LOH data in this analysis because we believe it is a finding common in CNC patients (see below). Furthermore, the increase in power expected from the use of LOH in combination with germ-line allelotyping for LOD score analysis, "depends on the certainty with which we can infer the phase between the cancer-predisposing allele and the marker allele retained in the tumor" (5). In the case of family A, Gadelha et al. (1) seem to have assigned the "disease-causing allele" for chromosome 2 to the unaffected father, with no apparent explanation. For chromosome 11, it was the maternal allele that was lost in the tumors, suggesting that, for this chromosome, the father was an unaffected carrier. However, the authors used their knowledge of genotyping information for 11 on their linkage analysis for chromosome 2, by assuming that in both cases it was the father who transmitted the disease-causing allele. Even if we accept the inclusion of LOH data for 11q13 calculations, the above assumption cannot be made for the 2p16 analysis. If the father in family A is considered as "unknown disease status" (UDS), which should be the case in the absence of any other information, the use of this kindred does not exclude 2p16, and the LOD scores become positive for the CNC locus (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 1. Maximum two-point LOD scores at {theta} = 0 (unless indicated otherwise) for family B under various assumptions (without incorporation of LOH data)1

 
Another issue is that of penetrance and affectation status for IFS; penetrance, for example, may not be the same in the two kindreds. Gadelha et al. (1) did not specify their age-related criteria for definition of the affectation status and penetrance of the disease was set at 0.9 for all calculations. However, the authors pointed out (and we agree) that the disease has "incomplete penetrance" (1). Because an age cut-off for penetrance was not given, an unaffected individual who carried the 11q13 "disease allele" (subject 13B) was considered to be UDS because of his age (11 yr). However, at least one of his siblings (subject 12B) was diagnosed with a tumor at the age of 13 yr (1, 2), and all of the affected members were diagnosed before 24 yr. We estimated the mean age of diagnosis of a pituitary tumor in the affected members of family B at 17.2 ± 4.14 yr (from the data given in Refs. 1 and 2). Taking into account that prospective screening would have identified potential abnormalities in GH secretion much earlier than the identification of a tumor, it is highly unlikely that individual 13B (now more than 12 yr old) is affected by IFS. Thus, 13B is either an unaffected carrier for IFS, in which case the penetrance of IFS in family B cannot be 0.9 [the maximum it could be is 0.71, because five of seven subjects who had the "disease-carrying" allele developed tumors, or he is truly unaffected, and IFS does not map to 11q13 because he has a recombination for the entire region.

Is there another somatotropinoma-causing tumor suppressor gene on 11q13? We have recently shown that LOH at 11q13 is present in CNC tumors, including a somatotropinoma (6). Other investigators, too, have shown that 11q13 LOH may occur in other tumors (4) and that most GH-producing adenomas that demonstrate LOH at 11q13 do not harbor somatic (tumor) mutations of menin (7).

In summary, although we cannot exclude entirely digenic inheritance or linkage to 11q13 (1), the two-point LOD scores and the presence of a possible recombination for 11q13 in family B lead to either inconclusive linkage or support that at least one of the families maps with greater likelihood to the CNC locus on 2p16 (Table 1Go). LOH at 11q13 is likely to be a tertiary hit at the tumor tissue level. In agreement with Gadelha et al. (1), there is much evidence to suggest that a tumor suppressor gene other than MEN 1 is located in 11q13, but we do not think that germ-line (vs. somatic) mutations of that gene are present in their families with IFS.

Received February 24, 2000.

References

  1. Gadelha MR, Une KN, Rohde K, Vaisman M, Kineman RD, Frohman LA. 2000 Isolated familial somatotropinomas: establishment of linkage to chromosome 11q13.1–11q13.3 and evidence for a potential second locus at chromosome 2p16-12. J Clin Endocrinol Metab. 85:707–714.[Abstract/Free Full Text]
  2. Gadelha MR, Prezant TR, Une KN, et al. 1999 Loss of heterozygosity on chromosome 11q13 in two families with acromegaly/gigantism is independent of mutations of the multiple endocrine neoplasia type I gene. J Clin Endocrinol Metab. 84:249–256.[Abstract/Free Full Text]
  3. Rohde K, Teare MD, Scherneck S, Santibanez-Koref M. 1995 A program using loss-of-constitutional-heterozygosity data to ascertain the location of predisposing genes in cancer families. Hum Hered. 45:337–345.[Medline]
  4. Nord B, Larsson C, Wong FK, Wallin G, Teh BT, Zedenius J. 1999 Sporadic follicular thyroid tumors show loss of a 200-kb region in 11q13 without evidence for mutations in the MEN1 gene. Genes Chromosomes Cancer. 26:35–39.[CrossRef][Medline]
  5. Rohde K, Teare MD, Santibanez Koref M. 1997Analysis of genetic linkage and somatic loss of heterozygosity in affected pairs of first-degree relatives. Am J Hum Genet. 61:418–422.
  6. Stratakis CA, Jenkins RB, Pras E, et al. 1996 Cytogenetic and microsatellite alterations in tumors from patients with the syndrome of myxomas, spotty skin pigmentation, and endocrine overactivity (Carney complex). J Clin Endocrinol Metab. 81:3607–3614.[Abstract]
  7. Zhuang Z, Ezzat SZ, Vortmeyer AO, et al. 1997 Mutations of the MEN1 tumor suppressor gene in pituitary tumors. Cancer Res. 57:5446–5451.[Abstract]