This study was conducted under the University of Utah IRB board 00028725, and patient consent was obtained. Patient and parental peripheral blood specimens were collected and DNA was extracted from blood leukocytes using the MagNa Pure Compact (Roche) procedure following the manufacture recommended protocol.

Samples were amplified by PCR for the coding regions and intron/exon boundaries of RET exons 10 - 11, and 13 - 16, then bidirectional Sanger sequencing was performed. Mutation Surveyor^® software (SoftGenetics, State College, PA) was used for data analysis.

Sample DNA was amplified by asymmetric PCR, highresolution melting (HRM) analysis was performed on the HR-1^TM (Idaho Technology, Salt Lake City, UT), and data were analyzed with custom software written in LabView (National Instruments, Austin TX) as described previously [8]. Briefly, after PCR, the unlabeled probe hybridizes to the amplicon template which is measured by LCgreen, a DNA intercalating dye. During HRM analysis, the amplicons with mismatches to the probe have a lower melting temperature (Tm) of dissociation than amplicons with complementary sequence to the probe [9].

The PCR primers used for RET exon 10, codon 620 interrogation were (forward 5’GGGCAGCATTGTTGGGGGAC3’ and reverse 5’TGGTGGTCCCGGCCGCCA3’). The unlabeled probes were 5’GGAGAAGTGCTTCTGCGAGCCCGAAGACATC3’ [wild type sequence at codon 620 (TGC)] and 5’GGAGAAGTGCTTCTTGGAGCCCGAAGACATC3’ [c.1859_1860delinsTG mutation sequence at codon 620 (TTG)]. Unlabeled probes have a blocking 3’amino modifier (IDT, Coralville, IA) to prevent extension during the PCR step.

In 2005, the patient presented at 19 years of age with a palpable thyroid nodule. The nodule was found to be 1.3 cm in the left thyroid lobe by ultrasound. The patient did not have dysphagia, voice changes, breathing issues, palpitations, previous radiation exposures, pain, or family history of MTC/MEN2. Fine needle aspiration showed neoplastic proliferation most consistent with medullary thyroid carcinoma (MTC). In concordance with the MTC diagnosis, calcitonin and carcinoembryonic antigen (CEA) levels were elevated: 1181 pg/mL and 53.3 ng/mL, respectively. The patient was screened for additional findings associated with MEN2, but had normal levels of parathyroid hormone, serum calcium, catecholamines, metanephrine and creatinine. Since the patient has no family history of endocrine problems or malignancy and the biochemical examination did not detect other findings of MEN2, sporadic MTC was suspected at the time of the surgery.

Total thyroidectomy with central compartment node dissection was preformed. A tumor of 1.2 × 1.1 × 1.0 cm was found in the left thyroid lobe. Multicentric or bilateral involvement was not identified. The tumor did not extend beyond the resection margins and capsular invasion was not identified. One of three lymph nodes tested positive for metastasis. The surgery confirmed the MTC diagnosis with small focus of perithyroidal connective tissue involvement and discovered metastasis to one of three perithyroidal lymph nodes. No unusual findings were found in the parathyroid tissue samples submitted for pathological examination. The left inferior parathyroid was autotransplanted to the left sternocleidomastoid. Calcitonin levels post-surgery remain borderline high at 13 - 15 pg/mL and were rising, suggesting the potential for recurrence.

The patient was tested at ARUP laboratories for germline RET mutations in exons 10, 11, and 13 - 16 by clinical Sanger sequencing (Figure 1(a)). Two heterozygous mutations were found, adjacent to each other within codon 620, a known hotspot for MEN2 causative RET mutations. Sanger sequencing of the parental samples, targeting RET exon 10 yielded only wild type sequence at codon 620. So, were these two RET mutations in trans c.[1859G > T;1860C > G] (p.[C620F;C620W]) or in cis c.1859_1860delinsTG (p.C620L)?

Since these two mutations affect adjacent base pairs, phase was determined using HRM and the unlabeled probe method [9] as described previously for RET mutation identification [8]. A probe complementary to the wild type RET sequence for codon 620 (TGC) was used (Figure 1(b)). Two control DNA samples each heterozygous for one codon 620 mutation [c.1859G > T (p.C620F) or c.1860C > G (p.C620W)] demonstrated a decreased Tm (melting temperature) for the mutation alleles compared to their wild type alleles. The patient’s sample demonstrated an allele with an even lower Tm than either of the control alleles harboring single nucleotide mutations. The patient sample also had an allele with the same Tm as the wild type control sample (which had only wild type alleles). In addition, a probe that complemented the c.1859_1860delinsTG mutation sequence at codon 620 (TTG) was used (Figure 1(c)). The patient’s DNA sequence complemented the probe resulting in a variant

allele with a higher Tm than either the wild type allele or the mutation alleles from the controls harboring single nucleotide mutations at codon 620. And again, the patient also had a wild type allele. The patient’s parents were Sanger sequenced at RET codon 620 and both had wild type sequence.