Abstract
Saturday, September 22, 2012
Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA
Thyroid Cancer Clinical 9:00 AM
Papillary thyroid carcinoma (PTC) is the fastest growing cancer type in the United States and many other countries. The increase in incidence mostly affects the follicular variant (FV) of PTC, whereas the proportion of classic-type PTC is decreasing. Although many FVPTC tumors are encapsulated and show indolent behavior, some are aggressive. RAS mutations are found in ∼40% of FVPTC, but driver mutations in the majority of FVPTC are unknown, limiting the ability to investigate the reasons for the increasing incidence and hampering the development of more individualized management of patients. Novel genetic tools, such as next-generation sequencing, can be used to discover new mutations in thyroid cancer.
Using real-time PCR and fluorescent melting-curve analysis and Sanger sequencing, we genotyped 501 consecutive PTCs for known point mutations, and TaqMan real-time PCR assays for known rearrangements. Nine mutation-negative FVPTCs and corresponding normal thyroid samples were studied using whole-genome sequencing (WGS) or whole-transcriptome sequencing (RNA-Seq) on the Illumina HiSeq 2000.
Of 456 tumors informative for analysis, 73% were positive for known mutations, whereas 27% were negative for known mutations tested, and the majority of those were FVPTC. Of those, 9 tumors with either indolent histopathologic features or metastatic behavior were subjected to WGS (3) and RNA-Seq (6). WGS had an average coverage of 49.3 reads and yielded 3–19 nonsynonymous single-nucleotide variations (SNVs) and 3–35 structural variations (SVs) per tumor. One metastatic FVPTC tumor revealed a fusion between TMP3 and NTRK1 with a novel breakpoint, and other tumors revealed several promising SNVs, including GLUT3 and MUC4. RNA-Seq data, analyzed using deFuse, yielded 7–29 fusion candidates for each tumor, including two fusions involving the thyroid genes TG and FOXE1/TTF2, which were involved in interchromosomal rearrangements. Validation of these results is ongoing.
Our next-generation sequencing analysis of FVPTC revealed a TPM3/NTRK1 fusion with a novel breakpoint, as well as several promising SNVs and SVs, which may allow better characterization of the biology and clinical behavior of FVPTC.
Alex & Simona Shnaider Laboratory in Molecular Oncology, Pathology & Laboratory Medicine, Mount Sinai Hospital, Toronto, ON, Canada
Thyroid Cancer Clinical 9:15 AM
We investigated the clinical significance of an Ep-ICD subcellular localization index (ESLI) in distinguishing aggressive papillary thyroid carcinoma (PTC) from nonaggressive cases using immunohistochemistry analyses.
Using domain-specific antibodies against the intracellular (Ep-ICD) and extracellular (EpEx) domains of epithelial cell adhesion molecule, 200 archived tissues from a new cohort of patients with benign thyroid disease as well as malignant aggressive and nonaggressive PTC were analyzed by immunohistochemistry (IHC). ESLI was defined as sum of the IHC scores for accumulation of nuclear (nuc) and cytoplasmic (cyt) Ep-ICD and loss of membranous EpEx, that is, ESLI = [Ep-ICDnuc + Ep-ICDcyt + loss of membranous EpEx].
For the benign thyroid tissues, nonaggressive PTC, and aggressive PTC, the mean ESLI scores were 4.5, 6.7, and 11, respectively. Immunofluorescence double staining confirmed increased nuclear Ep-ICD accumulation and decreased membrane EpEx expression in aggressive PTC. Receiver-operating characteristic (ROC) curve analysis showed an area under the curve (AUC) of 0.841, 70.2% sensitivity, and 83.9% specificity for nuclear Ep-ICD for differentiating aggressive PTC from nonaggressive PTC. ESLI distinguished aggressive PTC from nonaggressive cases with improved AUC of 0.924, 88.4% sensitivity, and 85.5% specificity. Our study confirms nuclear accumulation of Ep-ICD, and loss of membranous EpEx occurs in aggressive PTC, underscoring the potential of Ep-ICD and ESLI to serve as diagnostic markers for aggressive PTC. Kaplan–Meier survival analysis revealed significantly reduced disease-free survival (DFS) for ESLI-positive (cutoff >10) PTC (p<0.05), and mean DFS=133 months as compared to 210 months for patients who did not show positive ESLI.
ESLI scoring improves the identification of aggressive PTC, and thereby may serve as an useful index for defining aggressiveness and poor prognosis among PTC patients.
THYROGLOBULIN KNOCKOUT AND THE EFFECT OF ER STRESS RESPONSE ON PROTEOTOXIC THYROID CELL DEATH
Disorders of Thyroid Function Basic 9:30 AM
Most thyroidal protein synthesis is devoted to thyroglobulin (Tg), the precursor for T4 synthesis. Tg synthesis is physiologically linked to endoplasmic reticulum (ER) stress response. The Tg segment from linker to hinge within the Tg region I is rate limiting for Tg folding, requiring downstream regions II–III and ChEL for ER export to proceed. Both cog/cog mice and rdw/rdw rats have severe congenital hypothyroidism with a mutant Tg ChEL domain. Thyrocytes from both animals exhibit chronic ER stress (massive ER expansion with upregulated ER chaperones), yet confusingly, cog/cog mice develop a large goiter, whereas rdw/rdw rats get thyroid hypoplasia.
We made an rdw-Tg transgene driven by the bovine Tgn promoter; this rdw-Tg is expressed at much lower levels than endogenous Tg.
Thyroidal expression of mutant rdw-Tg in cog/cog mice does not improve hypothyroidism: the animals are runts (cretins) with serum TSH hugely elevated. However, rdw-Tg expression in cog/cog mice blocks goiter growth, by triggering thyroid cell death. The cell death is prevented by even one allele of wild-type (wt) Tg, which cross-dimerizes with rdw-Tg and rescues secretion. To determine if rdw-Tg proteotoxicity is triggered by ER stress, we have engineered the first Tgn-KO mouse. T4 supplementation allows these animals to grow and suppresses development of hypothyroid goiter. The animals were bred ±rdw-Tg transgene, which is induced when T4 supplementation is removed. In Tg-KO mice–even those induced for rdw-Tg–thyroidal ER chaperone levels are lower than those of normal mice. In spite of diminished overall ER stress, induced low-level expression of rdw-Tg in the absence of wt-Tg triggers a wave of thyroid cell death.
We conclude that the mutant gene product encoded by rdw (but not cog) kills thyrocytes when expressed in the absence of wt-Tg, and this is not necessarily triggered by excessive ER stress response, but by protein-specific proteotoxicity.
DEOB, NIDDK-NIH, Bethesda, MD
Thyroid Hormone Metabolism & Regulation Clinical 9:45 AM
Liothyronine (L-T3), the synthetic formulation of T3, is used in the treatment of thyroid cancer patients undergoing withdrawal for nuclear medicine procedures, and in some hypothyroid patients, in combination with levothyroxine (L-T4). An accurate estimation of the pharmacokinetics (PK) of L-T3 is important, since L-T3 has a narrow therapeutic index, and toxicity can result in serious adverse effects. The PK data of L-T3 are contradictory and mostly derived from single-dose measurements in healthy volunteers. Here, we present the PK of T3 during L-T3 withdrawal in the absence of endogenous thyroxine or exogenous L-T4.
Four thyroid cancer patients (2 women) age 55.7±9.6 years, undergoing TH withdrawal for clinically indicated nuclear medicine procedures, were studied. L-T3 (μg/μg) was substituted at a 1:3 ratio for L-T4 using a thrice-daily administration scheme. One month after L-T3 substitution, patients were admitted, and PK parameters were estimated after administration of last dose. Serum T3 levels were measured at 0, 30, 60, 90, 120, 150, 180, 210, 240, 360, 480, 600, and 720 minutes, and every 12 hours for the following ten days. TSH was measured every 12 hours. L-T3 PK parameters were derived using noncompartmental analysis. The average doses of L-T4 and L-T3 were 156.2±23.9 μg and 52.7±8.8 μg, respectively. Baseline serum T3 (normal range 90–215 ng/dL) and TSH (normal range 0.4–4.0 μIU/mL) levels were 194.7±50.5 ng/dL [range 162–270 ng/dL] and 0.01±0.01 μIU/mL, respectively. At day 11, after L-T3 withdrawal, TSH levels reached 47.5±19.0 μIU/mL. The T3 peak plasma concentration (Cmax) of 341.2±74.1 ng/dL was noted at 1.75±0.29 hours (Tmax), with an apparent half-life of 27.0±12.0 hours.
Our data indicate that the substitution of L-T3 for L-T4 at a 1:3 ratio on a thrice-daily administration schedule provides adequate suppressive therapy, while maintaining mean T3 concentrations within normal range. This modality results in TSH concentrations adequate for nuclear medicine therapy over a 10-day period, with minimal hypothyroid symptoms.
ClinicalTrials.gov identifier number:NCT01441154. Work supported by the NIDDK intramural research program Z01-DK047057-02.
Otolaryngology–Head and Neck Surgery, University of California, San Francisco, San Francisco, CA
Thyroid Development Basic 10:00 AM
The studies undertaken evaluate the potential of induced pluripotent stem (iPS) cells to differentiate into parathyroid-specific epithelial cells for autologous repair of tissue damage caused by disease and/or therapeutic intervention (e.g., surgery, chemotherapy, or radiation) and disease-related stem cells.
Directed differentiation of human iPS cells involved generation of embryoid bodies (EBs) followed by a 5-day treatment with Activin A, BMP4, FGF2, and LY294002 to enrich for definitive endoderm (DE). DE-enriched cells were exposed to NOGGIN and SB431542 for additional 2 days and then treated with Activin A, FGF10, KGF, BMP4, and WNT3A to direct the cells to generate anterior foregut endoderm (AFE)—the source of the parathyroid.
Cells were evaluated at different times for expression of differentiated stage-specific markers (i.e., c-KIT, EPCAM, Brachyury, MIXL1, CXCR4, SOX17, SOX2, FOXA2, FOXP2, PAX1, PAX9, PBX1, GCM2, CSR, CFTR, K18, and PTH). RT-PCR analysis of the RNA indicated progression of iPS cells to DE (days 1–5) and to AFE. Expression profiles showed enhanced expression of CXCR4 and FOXA2 and reduced SOX17 expression (DE markers) at day 5. Brachyury (mesoderm), MIXL1 (mesendoderm), SOX2 (self-renewal), and WNT3A (cell fate determination) expression was not detected. Day-7 cultures showed expression of SOX2 and FOXA2 (precursors of AFE). Cells were shown to express CSR, GCM2, and PT hallmarkers of parathyroid.
Endodermal differentiation of iPS cells can be achieved through modulation of transcriptional factors with growth factor, cytokines, and small molecules. DE and AEF provide the developmental basis for further differentiation into parathyroid and disease-related stem cells. Optimization is based on the prediction that culturing a population of PAX1+/PAX9+/PBX1+cells in the presence of BMP4, FGF8, and a low-dose Activin A before SHH induction will result in a higher proportion of parathyroid-like cells rather than thymic cells from same primordium.
Divison of Endocrinology and Nephrology, University of Leipzig, Leipzig, Germany
Thyroid Cancer Clinical 10:15 AM
Fine-needle aspiration (FNA) is the most sensitive method to select suspicious thyroid nodules for surgery. However, this method has inherent limitations, for example, indeterminate samples. As rearrangements (PAX8/PPARG and ET/PTC) and point mutations (BRAF, NRAS, HRAS, and KRAS) have been detected in follicular carcinomas (FTCs) and papillary carcinomas (PTCs), their detection in FNA smears could improve the diagnosis. However, rearrangements have up to date only been detected in fresh FNA material, and the number of FTCs was rather low in previous studies.
RNA and DNA were extracted from 310 routine air-dried FNA smears (164 indeterminate, 57 malignant, and 89 non-neoplastic) and corresponding formalin-fixed paraffin-embedded tissue (FFPE) samples (156 follicular adenomas, 32 FTCs, 9 follicular variant PTCs, 44 PTCs, and 69 goiters). PAX8/PPARG and RET/PTC1 and 3 rearrangements were detected by qPCR, whereas BRAF and RAS point mutations were detected by high-resolution melting (HRM)-PCR and by pyrosequencing.
On average, 8% and 3.9% of routine FNA samples did not allow analysis of a point mutation or rearrangement, respectively. For the 164 indeterminate samples, BRAF mutations could be detected in 1 FNA/1 FFPE sample, respectively and NRAS mutations in 12 FNA/21 FFPE samples, respectively. HRAS mutations were detected in 3 FNA/7 FFPE samples. No KRAS mutation was detected in the FNA samples. PAX8/PPARG was detected in 6 FNA/6 FFPE samples, whereas RET/PTC was not detected in any indeterminate sample. Molecular FNA screening increased the sensitivity from 67% (cytologyalone) to 75% (cytology+molecular FNA screening) in the total set. In the indeterminate set with 19 FTCs, the sensitivity of detecting carcinomas and mutation-positive adenomas was 48%, and specificity was 99%.
In summary, molecular screening for point mutations and rearrangements is feasible in routine air-dried FNA smears. Analyzing this panel of mutations, especially in indeterminate routine air-dried FNA smears, may reduce the number of diagnostic thyroid surgeries.
Short Call Abstract Author Index
Arvan, P. SCO-3
Boesenberg, E. SCO-6
Carty, S. SCO-1
Celi, F. SCO-4
Esmaeili-Shandiz, A. SCO-5
Eszlinger, M. SCO-6
Ferraz, C. SCO-6
Freeman, J. SCO-2
Gruenert, D. SCO-5
Gundelach, R. SCO-2
He, H. SCO-2
Hegedus, L. SCO-6
Hodak, S. SCO-1
Juarez, L. SCO-5
Kak, I. SCO-2
Kashat, L. SCO-2
Kellogg, A. SCO-3
Kelly, L. SCO-1
Kim, D. SCO-2
Koenig, R. SCO-3
Krogdahl, A. SCO-6
Kunavisarut, T. SCO-2
Lee, A. SCO-5
Linderman, J. SCO-4
Luo, J. SCO-1
MacMillan, C. SCO-2
Muenz, S. SCO-6
Nallani, S. SCO-4
Nikiforov, Y. SCO-1
Nikiforova, M. SCO-1
Orloff, L. SCO-5
Paschke, R. SCO-6
Peltsverger, M. SCO-4
Pengyuan, L. SCO-1
Precht Jensen, E. SCO-6
Pucino, F. SCO-4
Ralhan, R. SCO-2
Rehfeld, C. SCO-6
Sargent, R. G. SCO-5
Smith, S. SCO-4
So, A. SCO-2
Walfish, P. SCO-2
Wang, X. SCO-3
Yavuz, S. SCO-4
Yezzi, M. SCO-5