Purpose: Five germline genetic variants (rs116909374, rs965513, rs944289, rs966423, and rs2439302) have been associated in genome-wide association studies (GWAS) with increased risk of differentiated thyroid cancer (DTC), but their role in mortality of patients has not been established. Also, no preoperative marker of the clinical outcome of thyroid cancer had yet been identified. The aim of the study was to investigate the relationship between the variants and overall mortality in patients with DTC.

Experimental Design: Retrospective study of 1,836 patients (1,643 women, 193 men) with median age at diagnosis of 49 years and overall median follow-up time of 8.7 years after initial treatment at a single comprehensive cancer center between 1990 and 2013.

Results: Among 5 variants, rs966423 was associated with increased mortality, which was 6.4% (33 of 518) versus 3.7% (47 of 1,259) in TT carriers versus CC/CT carriers (P = 0.017). The HR of TT versus TC/CC carriers was 1.6 [95% confidence interval (CI), 1.02–2.49; P = 0.038] after adjustment for age at diagnosis and sex. Importantly, the association of rs966423 with mortality remained valid when clinicopathologic risk factors were included in the model (HR, 1.89; 95% CI, 1.14–3.13; P = 0.014). Higher rs966423–associated patient mortality of TT versus CC/CT carriers was also observed in interaction with angioinvasion (adjusted HR, 3.48; 95% CI, 1.67–7.22; P < 0.001), lymph node metastasis (adjusted HR, 3.47; 95% CI, 1.16–10.4; P = 0.018), extrathyroidal invasion (adjusted HR, 2.07; 95% CI, 1.15–3.73; P = 0.013).

Conclusions: The presence of the rs966423-TT genotype was associated with a significant increase in overall mortality of patients with DTC. Contrary to BRAF mutation and other somatic changes, the status of germline rs966423 is known before the treatment and might be used in the management of mortality risk by means of modification of therapy. Clin Cancer Res; 22(5); 1111–9. ©2015 AACR.

Translational Relevance

The study shows that a naturally occurring variant (rs966423) reported in genome-wide association studies (GWAS) as predisposing to thyroid cancer is associated with a significant increase in overall mortality of patients with thyroid cancer and importantly remains valid after adjustment for sex, age, lymph node metastasis, extrathyroidal invasion, angioinvasion, and distant metastasis [adjusted HR, 1.89; 95% confidence interval (CI), 1.14–3.13; P = 0.014]. The impact of the rs966423 variant is even more significant when combined with other clinicopathologic risk factors, for example, lymph node metastases (HR, 3.47; 95% CI, 1.16–10.4, P = 0.018). Contrary to BRAF mutation and other somatic changes putatively associated with patients' mortality, the status of germline rs966423 is known before the treatment and might be used in the management of mortality risk by means of modification of therapy; nevertheless, it requires further investigation in large prospective studies before it is ready for clinical application.

Differentiated thyroid carcinomas (DTC) constitute the most prevalent group of thyroid cancer. Two major histological subtypes: papillary (PTC) and follicular (FTC) thyroid carcinomas account for approximately 80% and 15% of all cases, respectively. DTC is generally considered a disease of low mortality as 90% of patients survive 10 years or more. Still, a significant number of patients die within first few years regardless of treatment, which usually consists of total or near total thyroidectomy, central or centrolateral neck lymphadenectomy, followed by thyroid ablation using low (30 mCi) or high (100–250 mCi) radioiodine (131I). All efforts are made to identify clinical or molecular markers characteristic for the patients with poor prognosis that would allow for offering them with a more aggressive therapy (1). Several factors have been described; unfortunately they are either of low effect like male sex and older age, or cannot be determined prior to the surgical treatment, like tumor angioinvasion, capsule invasion, or BRAF somatic mutation. Distant metastasis, the strongest indicator of poor prognosis, usually protrudes later. In a search of pretreatment clinical markers determining the poor prognosis of patients with thyroid cancer, we tested the impact of five inherited polymorphisms on overall survival. All five markers (rs116909374, rs965513, rs944289, rs966423, and rs2439302) have been previously shown by genome-wide association studies (GWAS) to increase the risk of differentiated thyroid cancer; however their impact on the clinical outcome has not been established (2–4).

Several large case–control studies, including hypothesis-free GWAS, have convincingly demonstrated genetic factors predisposing to DTCs (2–6). Still, there are no data showing the effect of SNPs on the survival rates of patients with thyroid cancer. Recent studies showed that somatic mutations are important predictors of the clinical outcome, as two nucleotide substitutions within the promoter of telomerase reverse transcriptase TERT gene were shown to occur with higher frequency in highly aggressive thyroid carcinomas (7, 8). Another recent study proposed the well-known BRAF V600E mutation as a determinant of the cancer-related mortality of patients with PTC (9), possibly due to its involvement in the radioiodine treatment response. However, the same study revealed that the significance of the association between BRAF V600E and PTC-related mortality is lost when the survival model includes other determinants like lymph node metastasis, extrathyroidal invasion, and distant metastases. Similarly, many other studies failed to prove such association (10).

In our previous study, we revealed a cumulative, additive effect of rs116909374, rs965513, rs944289, rs966423, and rs2439302 on the thyroid cancer risk (4). Because the influence of polymorphic variants on the overall survival has never been established, we analyzed the association of these SNPs with the mortality of patients with thyroid cancer. This is significant, as in contrary to somatic mutations, germline variants can be prognostic markers of clinical outcome prior to treatment.

Study patients

Patients eligible for the study included 1,836 patients from northern and central Poland, treated and followed up at the Maria Skłodowska-Curie Memorial Cancer Center-Institute of Oncology in Warsaw (Warsaw, Poland), during the years 1990–2013. Two pathologists, based on World Health Organization criteria, confirmed the diagnosis of DTC. Patient treatment consisted in a total or near-total thyroidectomy followed by radioactive iodine ablation and thyrotropin suppression. Follow-up or survival time indicates the time from the initial thyroidectomy to patient's death or, if the patient stayed alive, to the July 1, 2014, when the survival data were confirmed at the Polish Ministry of Internal Affairs. The demographic and clinical information, including pathologic characteristics and median follow-up time of specific subgroups of patients, are summarized in Table 1.

Table 1.

Demographic characteristics and follow-up time of patients

Age at diagnosis, median (interquartile range), yFollow-up, median (interquartile range), y
FactorNumber of patientsNumber of deaths (%)All patientsSurvivorsPatients who diedAll patientsSurvivorsPatients who died
All 1,836 85 (4.63) 48.58 (37.58–57.25) 47.92 (36.92–56.46) 65.92 (56.25–72.25) 8.67 (5–12.5) 8.75 (5.08–12.67) 6.75 (3.67–11.08) 
Gender 
 Female 1,643 70 (4.26) 48.58 (37.71–57) 48.08 (37.42–56.33) 65.92 (56.27–72.08) 8.75 (5.17–12.67) 8.83 (5.25–12.75) 7.12 (3.85–11.15) 
 Male 193 15 (7.77) 48.83 (35.58–58.83) 47 (35.04–57.96) 62.92 (56.33–74.62) 7.58 (4.25–11.58) 7.83 (4.33–11.9) 6.08 (3.04–9.88) 
Subtype 
 PTC cv 1,429 53 (3.71) 48.83 (37.67–57.08) 48.42 (37.31–56.5) 65.92 (56–73.33) 8.67 (5.08–12.5) 8.67 (5.17–12.5) 7.67 (4.08–11.83) 
 PTC fv 243 15 (6.17) 44.5 (35.79–55.29) 43.33 (34.58–53.58) 62.75 (55.21–71.21) 8.75 (4.75–11.92) 8.83 (4.81–11.9) 7.17 (2.83–11.29) 
 FTC 144 10 (6.94) 51 (38.9–60.52) 49.46 (38.1–58.65) 68.46 (62.79–72) 9.17 (4.25–14.67) 9.29 (4.33–15.04) 5.25 (3.54–8.71) 
 FTC insulare 20 7 (35) 58.33 (47.25–67.46) 62.58 (38.58–68) 56.67 (53.29–62.25) 5.71 (4.21–8.52) 6.67 (4.83–10.33) 2.75 (2.58–5.71) 
Stage 
 I 987 26 (2.63) 49.5 (39.71–56.75) 49.08 (39.33–56.5) 57.79 (53.56–67.52) 8.33 (4.83–11.67) 8.33 (4.83–11.67) 8.04 (4.6–11.83) 
 II 156 6 (3.85) 45.92 (37.21–57.02) 44.71 (36.21–55.83) 67.12 (66.48–73.52) 9.08 (6.4–13.33) 9.17 (6.75–13.46) 4.67 (3.29–7.42) 
 III 386 16 (4.15) 49.29 (34.77–59.1) 48.75 (34.5–58.31) 69.29 (56.6–72.02) 8.79 (4.69–12.5) 8.83 (4.69–12.65) 7.12 (4.73–10.77) 
 IV 232 33 (14.22) 45.67 (31.06–59.58) 40.75 (28.71–53.33) 67.83 (59.83–75.92) 8.38 (4.5–12) 9 (5.12–12.12) 5.75 (2.75–8.58) 
pT 
 pT1a 713 17 (2.38) 48.75 (39–56.17) 48.5 (38.92–56) 58.17 (55.17–66.08) 8.08 (4.75–11.42) 8.08 (4.75–11.42) 6.75 (3.92–10.58) 
 pT1b 411 13 (3.16) 47.67 (35.04–57.08) 47.38 (34.83–56.58) 65.92 (49.08–67.92) 8.58 (5.17–12.25) 8.67 (5.17–12.25) 7.67 (4.08–12) 
 pT2 194 8 (4.12) 45.25 (34.4–56.83) 44.29 (34–55.83) 67.12 (64.04–69.77) 8.96 (5.69–13.12) 9.08 (5.79–13.31) 5.71 (3.54–8.15) 
 pT3 372 31 (8.33) 49.96 (37.1–59.94) 48.83 (35.83–58.58) 69.08 (56.38–75.33) 8.58 (4.67–12.17) 8.92 (4.83–12.42) 5.17 (2.79–9.29) 
 pT4 47 9 (19.15) 57.08 (41.67–72.29) 50.92 (36.33–66.73) 72.33 (65.33–76.33) 8.67 (6.21–11) 9.17 (6.44–11.17) 7.08 (6.08–7.75) 
pN 
 N0 1,426 53 (3.72) 49.75 (39.92–57.48) 49.33 (39.58–57) 62.42 (55.17–69.83) 8.75 (5.33–12.67) 8.83 (5.33–12.75) 7.08 (4.42–11.08) 
 N1a 178 6 (3.37) 41.46 (29.71–54.06) 39.67 (29.27–53.35) 67.79 (65.96–74.88) 8.54 (4.33–11.5) 8.58 (4.4–11.52) 1.83 (1.33–8.9) 
 N1b 174 18 (10.34) 40.46 (28.04–54.6) 38.62 (27.21–49.98) 67.38 (57.54–76.23) 8.38 (4.75–12) 8.67 (5.08–12.23) 5.46 (3.35–8.9) 
pM 
 M0 1,728 65 (3.76) 48.42 (37.65–56.83) 47.92 (37.25–56.33) 65.33 (55.5–72.58) 8.79 (5.23–12.67) 8.83 (5.29–12.75) 7.67 (4.08–11.33) 
 M1 56 15 (26.79) 57.04 (36.92–67.15) 50.33 (31–59.92) 66.08 (62.04–69.54) 5.75 (3.65–9.15) 6.08 (4.17–10.25) 3.42 (2.75–6.58) 
Multifocality 
 s 1,279 59 (4.61) 48.42 (36.79–57.21) 47.67 (36–56.42) 64.17 (55.58–71.62) 8.75 (5.29–12.33) 8.83 (5.33–12.42) 6.67 (3.58–10.42) 
 m 473 17 (3.59) 49.58 (38.33–57.75) 48.71 (38.08–56.77) 68.42 (57.42–73.33) 8.08 (4.5–11.67) 8.08 (4.5–11.81) 4.83 (2.25–11.17) 
Size, mm 
 0–9 626 15 (2.4) 48.62 (38.77–56) 48.42 (38.71–55.62) 62.08 (55.58–68.96) 7.92 (4.83–11) 7.92 (4.83–11.04) 5.67 (3.38–9.75) 
 10–19 563 20 (3.55) 49.67 (37.62–58.5) 49.25 (37.21–57.58) 66.12 (56.94–72.33) 8.33 (4.79–11.58) 8.42 (4.83–11.58) 6.25 (3.23–8.31) 
 20–29 216 8 (3.7) 44.67 (32.42–57.08) 44 (31.69–56.15) 69.25 (48.38–75.67) 8.96 (5.81–12.69) 9 (5.9–12.79) 7.29 (4.21–10.25) 
 30–39 98 8 (8.16) 48.46 (38.15–60.65) 46.58 (36.42–56.33) 67.58 (64.48–75.1) 8.79 (5.15–12.65) 9.04 (5.15–13.19) 6.88 (4.94–8.4) 
 >40 159 21 (13.21) 50.33 (37.96–62.21) 48.38 (35.71–59.5) 64.17 (50.33–75.33) 8.58 (4.38–12.62) 8.92 (4.56–13.69) 5.17 (2.83–9.5) 
Angioinvasion 
 N 1,298 42 (3.24) 48.5 (38.08–56.58) 48.17 (37.67–56.25) 61.38 (55.25–69.75) 8.83 (5.5–12.65) 8.92 (5.5–12.69) 7.62 (4.33–11.04) 
 T 403 30 (7.44) 49.25 (34.12–59.88) 47.83 (33.42–58.58) 67.83 (58.21–75.19) 8.42 (4.38–11.5) 8.58 (4.5–11.75) 5.46 (2.5–9.27) 
Capsule 
 N 1,056 29 (2.75) 47.75 (37.4–56) 47.42 (36.92–55.83) 55.17 (47.5–66.08) 8.83 (5.31–12.75) 8.83 (5.33–12.75) 8.42 (4.58–12.17) 
 T0 361 20 (5.54) 49.58 (39–59.33) 48.67 (37.58–57.25) 66.62 (61.35–73.48) 8.08 (4.83–11.5) 8.25 (4.83–11.58) 7.17 (3.58–10.71) 
 T1 321 27 (8.41) 51.17 (37.33–61.17) 49.42 (36–59.17) 69.83 (61.92–76.5) 8.42 (4.83–11.5) 8.83 (5.08–11.71) 4.83 (2.58–7.46) 
rs944289 
 CC 259 12 (4.63) 48.42 (35.71–57.46) 47.17 (35.04–56.17) 67.38 (60.85–74.96) 8.92 (5.29–12.58) 9.08 (5.33–12.83) 8.17 (2.83–9.33) 
 CT 798 41 (5.14) 49 (38.58–57.06) 48.58 (38–56.42) 61.33 (54–69.83) 8.67 (5.08–12.5) 8.75 (5.17–12.58) 7.17 (3.92–12) 
 TT 705 27 (3.83) 47.33 (36.92–57.42) 46.88 (36.44–56.73) 67.33 (58.88–73.62) 8.33 (4.75–12.33) 8.42 (4.83–12.48) 6.08 (3.38–10.12) 
rs966423 
 TT 518 33 (6.37) 48.42 (37.35–57.06) 47.42 (36.58–55.75) 66.92 (56.33–72.58) 8.54 (5.1–12.94) 8.58 (5.25–13.08) 5.67 (2.83–8.92) 
 CT 883 32 (3.62) 48.67 (38.25–57.88) 48.42 (37.79–57.12) 64 (53.92–72.19) 8.75 (5.08–12.46) 8.83 (5.08–12.5) 7.58 (4.6–12.04) 
 CC 376 15 (3.99) 48.25 (36.27–56.54) 47.75 (36–56.08) 64.17 (55.79–70.88) 8.5 (4.73–12.27) 8.58 (4.75–12.33) 5.75 (3.58–7.79) 
rs2439302 
 CC 504 22 (4.37) 48.75 (36.79–58.17) 48.25 (35.1–57.06) 61.71 (56.48–70.52) 8.33 (4.67–11.79) 8.33 (4.69–11.73) 7.58 (4.92–11.77) 
 GC 878 38 (4.33) 48.5 (36.96–57.25) 47.92 (36.4–56.5) 66.92 (56.85–73.04) 8.67 (5–12.81) 8.75 (5–12.94) 6.88 (2.98–11.71) 
 GG 371 16 (4.31) 47.75 (38.79–56.33) 47.33 (38.5–55.83) 64.17 (51.98–72.92) 9 (5.12–12.42) 9 (5.46–12.5) 5.58 (3.19–9.79) 
rs116909374 
 CC 1,709 77 (4.51) 48.5 (37.5–57.33) 47.88 (36.73–56.5) 66.33 (56.67–72.33) 8.58 (4.92–12.42) 8.67 (5–12.5) 6.75 (3.42–11.08) 
 CT 108 7 (6.48) 49.04 (38.21–56.21) 48.42 (38.08–56.17) 54.17 (50.21–58.12) 9.38 (6.29–13.27) 9.67 (6.42–13.33) 6.33 (4.71–10.08) 
 TT 0 (0) 36.25 (36.25–36.25) 36.25 (36.25–36.25) NA 24 (24–24) 24 (24–24) NA 
rs965513 
 AA 372 14 (3.76) 47.42 (35.88–57.02) 47.12 (35.31–56.5) 64.04 (49.23–70.46) 9.08 (5.21–13.33) 9.21 (5.33–13.48) 5.75 (3.04–8.77) 
 AG 888 42 (4.73) 48.33 (37.81–56.5) 47.33 (37.27–55.75) 67.58 (59.85–74.67) 8.33 (4.83–12.33) 8.42 (4.83–12.33) 7.12 (3.46–11.96) 
 GG 518 23 (4.44) 50.08 (38.25–58) 49.58 (37.67–57.46) 59.25 (56.29–68.71) 8.58 (5.08–12.17) 8.67 (5.08–12.25) 7.58 (3.88–10.42) 
Age at diagnosis, median (interquartile range), yFollow-up, median (interquartile range), y
FactorNumber of patientsNumber of deaths (%)All patientsSurvivorsPatients who diedAll patientsSurvivorsPatients who died
All 1,836 85 (4.63) 48.58 (37.58–57.25) 47.92 (36.92–56.46) 65.92 (56.25–72.25) 8.67 (5–12.5) 8.75 (5.08–12.67) 6.75 (3.67–11.08) 
Gender 
 Female 1,643 70 (4.26) 48.58 (37.71–57) 48.08 (37.42–56.33) 65.92 (56.27–72.08) 8.75 (5.17–12.67) 8.83 (5.25–12.75) 7.12 (3.85–11.15) 
 Male 193 15 (7.77) 48.83 (35.58–58.83) 47 (35.04–57.96) 62.92 (56.33–74.62) 7.58 (4.25–11.58) 7.83 (4.33–11.9) 6.08 (3.04–9.88) 
Subtype 
 PTC cv 1,429 53 (3.71) 48.83 (37.67–57.08) 48.42 (37.31–56.5) 65.92 (56–73.33) 8.67 (5.08–12.5) 8.67 (5.17–12.5) 7.67 (4.08–11.83) 
 PTC fv 243 15 (6.17) 44.5 (35.79–55.29) 43.33 (34.58–53.58) 62.75 (55.21–71.21) 8.75 (4.75–11.92) 8.83 (4.81–11.9) 7.17 (2.83–11.29) 
 FTC 144 10 (6.94) 51 (38.9–60.52) 49.46 (38.1–58.65) 68.46 (62.79–72) 9.17 (4.25–14.67) 9.29 (4.33–15.04) 5.25 (3.54–8.71) 
 FTC insulare 20 7 (35) 58.33 (47.25–67.46) 62.58 (38.58–68) 56.67 (53.29–62.25) 5.71 (4.21–8.52) 6.67 (4.83–10.33) 2.75 (2.58–5.71) 
Stage 
 I 987 26 (2.63) 49.5 (39.71–56.75) 49.08 (39.33–56.5) 57.79 (53.56–67.52) 8.33 (4.83–11.67) 8.33 (4.83–11.67) 8.04 (4.6–11.83) 
 II 156 6 (3.85) 45.92 (37.21–57.02) 44.71 (36.21–55.83) 67.12 (66.48–73.52) 9.08 (6.4–13.33) 9.17 (6.75–13.46) 4.67 (3.29–7.42) 
 III 386 16 (4.15) 49.29 (34.77–59.1) 48.75 (34.5–58.31) 69.29 (56.6–72.02) 8.79 (4.69–12.5) 8.83 (4.69–12.65) 7.12 (4.73–10.77) 
 IV 232 33 (14.22) 45.67 (31.06–59.58) 40.75 (28.71–53.33) 67.83 (59.83–75.92) 8.38 (4.5–12) 9 (5.12–12.12) 5.75 (2.75–8.58) 
pT 
 pT1a 713 17 (2.38) 48.75 (39–56.17) 48.5 (38.92–56) 58.17 (55.17–66.08) 8.08 (4.75–11.42) 8.08 (4.75–11.42) 6.75 (3.92–10.58) 
 pT1b 411 13 (3.16) 47.67 (35.04–57.08) 47.38 (34.83–56.58) 65.92 (49.08–67.92) 8.58 (5.17–12.25) 8.67 (5.17–12.25) 7.67 (4.08–12) 
 pT2 194 8 (4.12) 45.25 (34.4–56.83) 44.29 (34–55.83) 67.12 (64.04–69.77) 8.96 (5.69–13.12) 9.08 (5.79–13.31) 5.71 (3.54–8.15) 
 pT3 372 31 (8.33) 49.96 (37.1–59.94) 48.83 (35.83–58.58) 69.08 (56.38–75.33) 8.58 (4.67–12.17) 8.92 (4.83–12.42) 5.17 (2.79–9.29) 
 pT4 47 9 (19.15) 57.08 (41.67–72.29) 50.92 (36.33–66.73) 72.33 (65.33–76.33) 8.67 (6.21–11) 9.17 (6.44–11.17) 7.08 (6.08–7.75) 
pN 
 N0 1,426 53 (3.72) 49.75 (39.92–57.48) 49.33 (39.58–57) 62.42 (55.17–69.83) 8.75 (5.33–12.67) 8.83 (5.33–12.75) 7.08 (4.42–11.08) 
 N1a 178 6 (3.37) 41.46 (29.71–54.06) 39.67 (29.27–53.35) 67.79 (65.96–74.88) 8.54 (4.33–11.5) 8.58 (4.4–11.52) 1.83 (1.33–8.9) 
 N1b 174 18 (10.34) 40.46 (28.04–54.6) 38.62 (27.21–49.98) 67.38 (57.54–76.23) 8.38 (4.75–12) 8.67 (5.08–12.23) 5.46 (3.35–8.9) 
pM 
 M0 1,728 65 (3.76) 48.42 (37.65–56.83) 47.92 (37.25–56.33) 65.33 (55.5–72.58) 8.79 (5.23–12.67) 8.83 (5.29–12.75) 7.67 (4.08–11.33) 
 M1 56 15 (26.79) 57.04 (36.92–67.15) 50.33 (31–59.92) 66.08 (62.04–69.54) 5.75 (3.65–9.15) 6.08 (4.17–10.25) 3.42 (2.75–6.58) 
Multifocality 
 s 1,279 59 (4.61) 48.42 (36.79–57.21) 47.67 (36–56.42) 64.17 (55.58–71.62) 8.75 (5.29–12.33) 8.83 (5.33–12.42) 6.67 (3.58–10.42) 
 m 473 17 (3.59) 49.58 (38.33–57.75) 48.71 (38.08–56.77) 68.42 (57.42–73.33) 8.08 (4.5–11.67) 8.08 (4.5–11.81) 4.83 (2.25–11.17) 
Size, mm 
 0–9 626 15 (2.4) 48.62 (38.77–56) 48.42 (38.71–55.62) 62.08 (55.58–68.96) 7.92 (4.83–11) 7.92 (4.83–11.04) 5.67 (3.38–9.75) 
 10–19 563 20 (3.55) 49.67 (37.62–58.5) 49.25 (37.21–57.58) 66.12 (56.94–72.33) 8.33 (4.79–11.58) 8.42 (4.83–11.58) 6.25 (3.23–8.31) 
 20–29 216 8 (3.7) 44.67 (32.42–57.08) 44 (31.69–56.15) 69.25 (48.38–75.67) 8.96 (5.81–12.69) 9 (5.9–12.79) 7.29 (4.21–10.25) 
 30–39 98 8 (8.16) 48.46 (38.15–60.65) 46.58 (36.42–56.33) 67.58 (64.48–75.1) 8.79 (5.15–12.65) 9.04 (5.15–13.19) 6.88 (4.94–8.4) 
 >40 159 21 (13.21) 50.33 (37.96–62.21) 48.38 (35.71–59.5) 64.17 (50.33–75.33) 8.58 (4.38–12.62) 8.92 (4.56–13.69) 5.17 (2.83–9.5) 
Angioinvasion 
 N 1,298 42 (3.24) 48.5 (38.08–56.58) 48.17 (37.67–56.25) 61.38 (55.25–69.75) 8.83 (5.5–12.65) 8.92 (5.5–12.69) 7.62 (4.33–11.04) 
 T 403 30 (7.44) 49.25 (34.12–59.88) 47.83 (33.42–58.58) 67.83 (58.21–75.19) 8.42 (4.38–11.5) 8.58 (4.5–11.75) 5.46 (2.5–9.27) 
Capsule 
 N 1,056 29 (2.75) 47.75 (37.4–56) 47.42 (36.92–55.83) 55.17 (47.5–66.08) 8.83 (5.31–12.75) 8.83 (5.33–12.75) 8.42 (4.58–12.17) 
 T0 361 20 (5.54) 49.58 (39–59.33) 48.67 (37.58–57.25) 66.62 (61.35–73.48) 8.08 (4.83–11.5) 8.25 (4.83–11.58) 7.17 (3.58–10.71) 
 T1 321 27 (8.41) 51.17 (37.33–61.17) 49.42 (36–59.17) 69.83 (61.92–76.5) 8.42 (4.83–11.5) 8.83 (5.08–11.71) 4.83 (2.58–7.46) 
rs944289 
 CC 259 12 (4.63) 48.42 (35.71–57.46) 47.17 (35.04–56.17) 67.38 (60.85–74.96) 8.92 (5.29–12.58) 9.08 (5.33–12.83) 8.17 (2.83–9.33) 
 CT 798 41 (5.14) 49 (38.58–57.06) 48.58 (38–56.42) 61.33 (54–69.83) 8.67 (5.08–12.5) 8.75 (5.17–12.58) 7.17 (3.92–12) 
 TT 705 27 (3.83) 47.33 (36.92–57.42) 46.88 (36.44–56.73) 67.33 (58.88–73.62) 8.33 (4.75–12.33) 8.42 (4.83–12.48) 6.08 (3.38–10.12) 
rs966423 
 TT 518 33 (6.37) 48.42 (37.35–57.06) 47.42 (36.58–55.75) 66.92 (56.33–72.58) 8.54 (5.1–12.94) 8.58 (5.25–13.08) 5.67 (2.83–8.92) 
 CT 883 32 (3.62) 48.67 (38.25–57.88) 48.42 (37.79–57.12) 64 (53.92–72.19) 8.75 (5.08–12.46) 8.83 (5.08–12.5) 7.58 (4.6–12.04) 
 CC 376 15 (3.99) 48.25 (36.27–56.54) 47.75 (36–56.08) 64.17 (55.79–70.88) 8.5 (4.73–12.27) 8.58 (4.75–12.33) 5.75 (3.58–7.79) 
rs2439302 
 CC 504 22 (4.37) 48.75 (36.79–58.17) 48.25 (35.1–57.06) 61.71 (56.48–70.52) 8.33 (4.67–11.79) 8.33 (4.69–11.73) 7.58 (4.92–11.77) 
 GC 878 38 (4.33) 48.5 (36.96–57.25) 47.92 (36.4–56.5) 66.92 (56.85–73.04) 8.67 (5–12.81) 8.75 (5–12.94) 6.88 (2.98–11.71) 
 GG 371 16 (4.31) 47.75 (38.79–56.33) 47.33 (38.5–55.83) 64.17 (51.98–72.92) 9 (5.12–12.42) 9 (5.46–12.5) 5.58 (3.19–9.79) 
rs116909374 
 CC 1,709 77 (4.51) 48.5 (37.5–57.33) 47.88 (36.73–56.5) 66.33 (56.67–72.33) 8.58 (4.92–12.42) 8.67 (5–12.5) 6.75 (3.42–11.08) 
 CT 108 7 (6.48) 49.04 (38.21–56.21) 48.42 (38.08–56.17) 54.17 (50.21–58.12) 9.38 (6.29–13.27) 9.67 (6.42–13.33) 6.33 (4.71–10.08) 
 TT 0 (0) 36.25 (36.25–36.25) 36.25 (36.25–36.25) NA 24 (24–24) 24 (24–24) NA 
rs965513 
 AA 372 14 (3.76) 47.42 (35.88–57.02) 47.12 (35.31–56.5) 64.04 (49.23–70.46) 9.08 (5.21–13.33) 9.21 (5.33–13.48) 5.75 (3.04–8.77) 
 AG 888 42 (4.73) 48.33 (37.81–56.5) 47.33 (37.27–55.75) 67.58 (59.85–74.67) 8.33 (4.83–12.33) 8.42 (4.83–12.33) 7.12 (3.46–11.96) 
 GG 518 23 (4.44) 50.08 (38.25–58) 49.58 (37.67–57.46) 59.25 (56.29–68.71) 8.58 (5.08–12.17) 8.67 (5.08–12.25) 7.58 (3.88–10.42) 

Study design

The study was conducted on the basis of genotyping data and retrospective clinical data of the patients treated for thyroid cancer in a single comprehensive cancer center, the Maria Skłodowska-Curie Memorial Cancer Center-Institute of Oncology in Warsaw. The Institutional Review Board has approved the study and all subjects gave written informed consent before participation. Genotyping of the five analyzed SNPs (rs116909374, rs965513, rs944289, rs966423, rs2439302) in blood-derived DNA of patient with DTC was performed using the Sequenom technology, as described previously (4), and clinicopathologic data were collected during medical examinations as well as from medical archives.

Statistical analyses

Patient mortality was determined by dividing the number of deaths by the number of patients in each analyzed group. The mortality rates within each genotype were compared using the Fisher exact test. The number of deaths was divided by the sum of total follow-up years to produce the rates per person-year, and the 95% confidence intervals (CI) were calculated using the Poisson regression. The general genotype-dependent survivals were estimated and compared using the Kaplan–Meier survival curves, log-rank tests, and Cox proportional hazard regression analyses that censored patients at the time of the database inquire (July 1, 2014) or the follow-up period of 20 years. Proportional hazards regressions were adjusted by stratification for age at diagnosis (<45 and ≥45 years old) and sex using the counting process formulation of Andersen and Gill (11). Subgroup analyses were not adjusted for multiple comparisons and should be considered exploratory. All P values are 2-sided and significance was set at P < 0.05. All statistical analyses were performed using the R software package version 3.0.2.

Relationship between 5 GWAS-derived variants and overall mortality

The study included 1,836 patients (1,643 women and 193 men) with a median age at diagnosis of 48.6 years (interquartile range, 38–57 years) and an overall median follow-up time of 8.7 years (interquartile range, 5–12.5 years). The frequencies of each genotype in the 5 GWAS variants: rs116909374, rs965513, rs944289, rs966423, and rs2439302 are provided in Table 1. The overall mortality of all cases was 4.63% (85 of 1,836). There were 53 deaths among cases with conventional variant of PTC (3.71%), 15 deaths among follicular variant of PTC (6.17%), 10 deaths among patients with FTC (6.94%), and 7 with insular FTC (35%). Mortality percentages and deaths per 1,000 person-years for all 5 genetic variants are reported in Table 1.

Among the five variants, rs966423 was associated with increased mortality reaching 6.4% (33 of 518) in TT-carriers versus 3.7% (47 of 1,259) in CC- or CT-carriers (P = 0.017; Table 2). Deaths per 1,000 person-years were 6.81 (95% CI, 4.69–9.56) in TT-carriers versus 4.01 (95% CI, 2.95–5.34) in CC/CT-carriers; the HR was 1.6 (95% CI, 1.02–2.49; P = 0.038) after adjustment for age at diagnosis and sex. After dividing patients according to the histopathologic variants with lower numbers of cases and deaths in specific subgroups, no significant results were observed (Supplementary Table S1). For the largest subgroup consisting of PTC cases, the mortality was 5.3% (25 of 468) in TT-carriers versus 3.3% (38 of 1,146) in CC/CT-carriers (P = 0.065).

Table 2.

Overall mortality, person-years of follow-up, and HRs for risk genotypes

Mortality, n/total (%)Deaths per 1,000 person-years (95% CI)UnadjustedAdjusted
SNPOverallRisk homozygotesHeterozygotes/homozygotesPPerson-years follow-upRisk homozygotesHeterozygotes/homozygotesLog-rank PHR (95%CI)Log-rank PHR (95%CI)
rs944289 (CC vs. CT/TT) 80/1,762 (4.5) 12/259 (4.6) 68/1,503 (4.5) 0.873 16,413 (13,931 ± 2,482) 4.83 (2.5–8.44) 4.88 (3.79–6.19) 0.986 0.99 (0.54–1.84) 0.751 1.11 (0.6–2.05) 
rs966423 (TT vs. CT/CC) 80/1,777 (4.5) 33/518 (6.4) 47/1,259 (3.7) 0.017 16,557 (11,709 ± 4,848) 6.81 (4.69–9.56) 4.01 (2.95–5.34) 0.022 1.67 (1.07–2.61) 0.038 1.6 (1.02–2.49) 
rs2439302 (CC vs. GC/GG) 76/1,753 (4.3) 22/504 (4.4) 54/1,249 (4.3) 16,268 (11,774 ± 4,494) 4.9 (3.07–7.41) 4.59 (3.45–5.98) 0.709 1.1 (0.67–1.81) 0.792 1.07 (0.65–1.76) 
rs116909374 (CC vs. CT/TT) 84/1,818 (4.6) 77/1,709 (4.5) 7/109 (6.4) 0.343 17,019 (1,112 ± 15,907) 4.84 (3.82–6.05) 6.29 (2.53–12.97) 0.572 0.8 (0.37–1.74) 0.674 0.85 (0.39–1.84) 
rs965513 (GG vs. GA/AA) 79/1,778 (4.4) 23/518 (4.4) 56/1,260 (4.4) 16,594 (11,874 ± 4,720) 4.87 (3.09–7.31) 4.72 (3.56–6.12) 0.781 1.07 (0.66–1.75) 0.813 1.06 (0.65–1.73) 
Mortality, n/total (%)Deaths per 1,000 person-years (95% CI)UnadjustedAdjusted
SNPOverallRisk homozygotesHeterozygotes/homozygotesPPerson-years follow-upRisk homozygotesHeterozygotes/homozygotesLog-rank PHR (95%CI)Log-rank PHR (95%CI)
rs944289 (CC vs. CT/TT) 80/1,762 (4.5) 12/259 (4.6) 68/1,503 (4.5) 0.873 16,413 (13,931 ± 2,482) 4.83 (2.5–8.44) 4.88 (3.79–6.19) 0.986 0.99 (0.54–1.84) 0.751 1.11 (0.6–2.05) 
rs966423 (TT vs. CT/CC) 80/1,777 (4.5) 33/518 (6.4) 47/1,259 (3.7) 0.017 16,557 (11,709 ± 4,848) 6.81 (4.69–9.56) 4.01 (2.95–5.34) 0.022 1.67 (1.07–2.61) 0.038 1.6 (1.02–2.49) 
rs2439302 (CC vs. GC/GG) 76/1,753 (4.3) 22/504 (4.4) 54/1,249 (4.3) 16,268 (11,774 ± 4,494) 4.9 (3.07–7.41) 4.59 (3.45–5.98) 0.709 1.1 (0.67–1.81) 0.792 1.07 (0.65–1.76) 
rs116909374 (CC vs. CT/TT) 84/1,818 (4.6) 77/1,709 (4.5) 7/109 (6.4) 0.343 17,019 (1,112 ± 15,907) 4.84 (3.82–6.05) 6.29 (2.53–12.97) 0.572 0.8 (0.37–1.74) 0.674 0.85 (0.39–1.84) 
rs965513 (GG vs. GA/AA) 79/1,778 (4.4) 23/518 (4.4) 56/1,260 (4.4) 16,594 (11,874 ± 4,720) 4.87 (3.09–7.31) 4.72 (3.56–6.12) 0.781 1.07 (0.66–1.75) 0.813 1.06 (0.65–1.73) 

The independent association of clinical factors and rs966423 with increased overall mortality

We observed a significantly higher risk of death associated with several factors that are known to increase the mortality of patients with thyroid cancer, including male gender, age older than 45 years, tumor size > 30 mm, lymph node metastases, distant metastases, angioinvasion, and extrathyroidal invasion (Fig. 1). In the multivariate model, comprising such factors as sex, age, lymph node metastasis, distant metastasis, extrathyroidal invasion, angioinvasion, and rs966423, the increased mortality was independently associated with only four factors: age older than 45 years (HR, 11.97; 95% CI, 4.73–30.26; Wald P = 1.56e-07), extrathyroidal invasion (HR, 1.8; 95% CI, 1.03–3.16; Wald P = 0.04), distant metastasis (HR, 9.24; 95% CI, 4.57–18.66; Wald P = 5.85e-10), and rs966423 (HR, 1.89; 95% CI, 1.14–3.13; Wald P = 0.014). The association was not significant for sex (P = 0.674), lymph nodes metastases (P = 0.228), and angioinvasion (P = 0.054; Table 3).

Figure 1.

Stratified analysis of various clinicopathologic factors associated with survival outcome of patients with thyroid cancer. Vertical line represents the HR equal to 1.

Figure 1.

Stratified analysis of various clinicopathologic factors associated with survival outcome of patients with thyroid cancer. Vertical line represents the HR equal to 1.

Close modal
Table 3.

Multivariate analysis of genetic and clinicopathologic risk factors

No.FactorWald PHR (95%CI)
rs966423 (TT) 0.014 1.89 (1.14–3.13) 
Sex (M) 0.674 1.17 (0.57–2.38) 
Age (≥45 y) 1.56E-07 11.97 (4.73–30.26) 
Angioinvasion 0.054 1.76 (0.99–3.13) 
Extrathyroidal invasion 0.04 1.8 (1.03–3.16) 
Lymph node mets (N1b) 0.228 1.56 (0.76–3.2) 
Distant metastasis (M1) 5.85E-10 9.24 (4.57–18.66) 
No.FactorWald PHR (95%CI)
rs966423 (TT) 0.014 1.89 (1.14–3.13) 
Sex (M) 0.674 1.17 (0.57–2.38) 
Age (≥45 y) 1.56E-07 11.97 (4.73–30.26) 
Angioinvasion 0.054 1.76 (0.99–3.13) 
Extrathyroidal invasion 0.04 1.8 (1.03–3.16) 
Lymph node mets (N1b) 0.228 1.56 (0.76–3.2) 
Distant metastasis (M1) 5.85E-10 9.24 (4.57–18.66) 

Interaction between rs966423 and clinical factors in shaping the overall mortality

Rs966423 modifies the mortality risk associated with several conventional clinicopathologic risk factors, including age, angioinvasion, extrathyroidal invasion, and lymph node metastasis (Table 4). Importantly, the HRs remained statistically significant also after adjustment for age and sex. The differences in overall mortality in patients with the rs966423 risk genotype coexisting with various clinicopathologic factors were reflected in the Kaplan–Meier survival curves (Fig. 2A).

Figure 2.

Kaplan–Meier survival curves of the interaction of rs966423 variant with clinicopathological risk factors (A) and age (B) in affecting overall survival of patients with thyroid cancer. In all panels, follow-up time is truncated at 20 years. In each panel, P values are from the log-rank test.

Figure 2.

Kaplan–Meier survival curves of the interaction of rs966423 variant with clinicopathological risk factors (A) and age (B) in affecting overall survival of patients with thyroid cancer. In all panels, follow-up time is truncated at 20 years. In each panel, P values are from the log-rank test.

Close modal
Table 4.

Overall mortality, person-years of follow-up, and HRs for rs966423 TT carriers and CT/CC carriers in various clinicopathologic categories

Mortality, n/total (%)Deaths per 1,000 person-years (95% CI)UnadjustedAdjusted
CategoryOverallrs966423-TTrs966423-CC/CTPPerson-years follow-uprs966423-TTrs966423-CC/CTLog-rank PHR (95%CI)Log-rank PHR (95% CI)
All patients 80/1,777 (4.5) 33/518 (6.4) 47/1,259 (3.7) 0.017 16,557 (11,709 ± 4,848) 6.81 (4.69–9.56) 4.01 (2.95–5.34) 0.022 1.67 (1.07–2.61) 0.038 1.6 (1.02–2.49) 
Gender 
 Female 66/1,586 (4.2) 26/454 (5.7) 40/1,132 (3.5) 0.052 14,913 (10,578 ± 4,335) 6 (3.92–8.79) 3.78 (2.7–5.15) 0.076 1.56 (0.95–2.55) 0.12 1.48 (0.9–2.42) 
 Male 14/191 (7.3) 7/64 (10.9) 7/127 (5.5) 0.238 1,644 (1,130 ± 514) 13.63 (5.48–28.08) 6.19 (2.49–12.76) 0.126 2.22 (0.78–6.35) 0.115 2.28 (0.79–6.57) 
Age_45 
 <45 7/741 (0.9) 2/214 (0.9) 5/527 (0.9) 7,325 (5,248 ± 2,077) 0.96 (0.12–3.48) 0.95 (0.31–2.22) 0.992 0.99 (0.19–5.11) 0.99 0.99 (0.19–5.1) 
 ≥45 73/1,036 (7) 31/304 (10.2) 42/732 (5.7) 0.016 9,232 (6,460 ± 2,772) 11.18 (7.6–15.88) 6.5 (4.69–8.79) 0.028 1.67 (1.05–2.67) 0.03 1.66 (1.04–2.65) 
Age_45_65 
 <45 7/741 (0.9) 2/214 (0.9) 5/527 (0.9) 7,325 (5,248 ± 2,077) 0.96 (0.12–3.48) 0.95 (0.31–2.22) 0.992 0.99 (0.19–5.11) 0.99 0.99 (0.19–5.1) 
 45–65 31/825 (3.8) 12/242 (5) 19/583 (3.3) 0.235 7,645 (5,348 ± 2,298) 5.22 (2.7–9.12) 3.55 (2.14–5.55) 0.356 1.4 (0.68–2.9) 0.357 1.41 (0.68–2.91) 
 >65 42/211 (19.9) 19/62 (30.6) 23/149 (15.4) 0.014 1,586 (1,112.5 ± 474) 40.08 (24.13–62.6) 20.67 (13.11–31.02) 0.035 1.91 (1.04–3.5) 0.038 1.89 (1.03–3.47) 
pT 
 pT1 28/1,085 (2.6) 9/298 (3) 19/787 (2.4) 0.668 9,514 (6,792 ± 2,722) 3.31 (1.51–6.28) 2.8 (1.68–4.37) 0.775 1.12 (0.51–2.49) 0.78 1.13 (0.49–2.59) 
 pT2 8/189 (4.2) 2/56 (3.6) 6/133 (4.5) 1,777 (1,278 ± 499) 4.01 (0.49–14.48) 4.69 (1.72–10.22) 0.824 0.83 (0.17–4.14) 0.576 0.63 (0.13–3.16) 
 pT3 29/362 (8) 17/118 (14.4) 12/244 (4.9) 0.003 3,265 (2,237.5 ± 1,028) 16.54 (9.63–26.48) 5.36 (2.77–9.37) 0.002 3.05 (1.46–6.4) 0.01 2.61 (1.23–5.53) 
 pT4 9/46 (19.6) 3/19 (15.8) 6/27 (22.2) 0.716 404 (241 ± 163) 18.38 (3.79–53.7) 24.9 (9.14–54.21) 0.733 0.79 (0.19–3.17) 0.775 0.81 (0.19–3.42) 
pN 
 N0 51/1,383 (3.7) 16/375 (4.3) 35/1,008 (3.5) 0.521 13,098 (9,440 ± 3,658) 4.37 (2.5–7.1) 3.71 (2.58–5.16) 0.649 1.15 (0.63–2.07) 0.781 1.09 (0.6–1.98) 
 N1a 6/171 (3.5) 4/65 (6.2) 2/106 (1.9) 0.202 1,492 (941 ± 551) 7.26 (1.98–18.6) 2.13 (0.26–7.68) 0.145 3.29 (0.6–17.97) 0.148 3.29 (0.6–18.1) 
 N1b 15/166 (9) 10/59 (16.9) 5/107 (4.7) 0.011 1,492 (987 ± 505) 19.8 (9.5–36.42) 5.07 (1.65–11.82) 0.008 3.84 (1.31–11.24) 0.018 3.47 (1.16–10.4) 
pM 
 M0 60/1,670 (3.6) 25/479 (5.2) 35/1,191 (2.9) 0.029 15,763 (11,172 ± 4,591) 5.45 (3.52–8.04) 3.13 (2.18–4.36) 0.04 1.7 (1.02–2.84) 0.065 1.62 (0.97–2.71) 
 M1 15/55 (27.3) 6/21 (28.6) 9/34 (26.5) 357 (226 ± 131) 45.86 (16.83–99.82) 39.78 (18.19–75.51) 0.775 1.16 (0.41–3.27) 0.198 1.97 (0.69–5.6) 
Capsule 
 N 27/1,024 (2.6) 19/750 (2.5) 8/274 (2.9) 0.826 9,780 (7,025.5 ± 2,754) 2.7 (1.63–4.22) 2.9 (1.25–5.72) 0.894 1.06 (0.46–2.42) 0.861 0.92 (0.38–2.22) 
 T0 20/347 (5.8) 12/234 (5.1) 8/113 (7.1) 0.469 2,989 (2,055 ± 934) 5.84 (3.02–10.2) 8.57 (3.7–16.88) 0.389 1.48 (0.6–3.62) 0.356 1.52 (0.62–3.72) 
 T1 25/312 (8) 11/211 (5.2) 14/101 (13.9) 0.013 2,739.5 (1,879 ± 861) 5.85 (2.92–10.48) 16.26 (8.89–27.29) 0.007 2.83 (1.28–6.24) 0.011 2.73 (1.22–6.11) 
Angioinvasion 
 N 38/1,253 (3) 12/362 (3.3) 26/891 (2.9) 0.718 11,939 (8,401 ± 3,538) 3.39 (1.75–5.92) 3.09 (2.02–4.53) 0.806 1.09 (0.55–2.16) 0.881 0.95 (0.47–1.93) 
 T 30/394 (7.6) 17/117 (14.5) 13/277 (4.7) 0.001 3,401 (2,441 ± 960) 17.71 (10.31–28.35) 5.33 (2.84–9.11) 0.001 3.28 (1.59–6.76) 3.48 (1.67–7.22) 
Multifocality 
 s 55/1,235 (4.5) 21/370 (5.7) 34/865 (3.9) 0.178 11,479 (8,065 ± 3,414) 6.15 (3.81–9.4) 4.22 (2.92–5.89) 0.157 1.48 (0.86–2.55) 0.161 1.48 (0.85–2.56) 
 m 17/459 (3.7) 8/125 (6.4) 9/334 (2.7) 0.091 3,921 (2,828 ± 1,093) 7.32 (3.16–14.42) 3.18 (1.46–6.04) 0.169 1.97 (0.73–5.3) 0.29 1.7 (0.63–4.61) 
Size, mm 
 0–9 15/606 (2.5) 6/153 (3.9) 9/453 (2) 0.226 5,052 (3,752 ± 1,300) 4.62 (1.69–10.05) 2.4 (1.1–4.55) 0.222 1.88 (0.67–5.29) 0.142 2.17 (0.75–6.26) 
 10–19 18/544 (3.3) 9/173 (5.2) 9/371 (2.4) 0.121 4,770 (3,187 ± 1,583) 5.69 (2.6–10.79) 2.82 (1.29–5.36) 0.235 1.77 (0.68–4.59) 0.216 1.81 (0.7–4.72) 
 20–29 7/212 (3.3) 0/59 (0) 7/153 (4.6) 0.194 2,016 (1,450 ± 566) 0 (0–6.52) 4.83 (1.94–9.94) 0.095 0 (0-Inf) 0.063 0 (0-Inf) 
 30–39 8/95 (8.4) 4/35 (11.4) 4/60 (6.7) 0.461 861 (571 ± 290) 13.78 (3.75–35.28) 7 (1.91–17.93) 0.304 2.04 (0.51–8.18) 0.342 1.94 (0.48–7.78) 
 >40 21/153 (13.7) 10/48 (20.8) 11/105 (10.5) 0.126 1,387 (991 ± 396) 25.26 (12.11–46.46) 11.1 (5.54–19.86) 0.064 2.2 (0.93–5.19) 0.2 1.75 (0.74–4.16) 
Stage 
 I 24/956 (2.5) 8/251 (3.2) 16/705 (2.3) 0.481 8,437 (6,120 ± 2,317) 3.45 (1.49–6.8) 2.61 (1.49–4.25) 0.632 1.23 (0.52–2.89) 0.722 1.18 (0.48–2.89) 
 II 6/151 (4) 1/45 (2.2) 5/106 (4.7) 0.67 1,465 (1,042 ± 423) 2.37 (0.06–13.18) 4.8 (1.56–11.2) 0.484 0.47 (0.06–4.05) 0.384 0.4 (0.05–3.42) 
 III 16/374 (4.3) 8/124 (6.5) 8/250 (3.2) 0.175 3,461 (2,346 ± 1,114) 7.18 (3.1–14.15) 3.41 (1.47–6.72) 0.137 2.07 (0.78–5.52) 0.208 1.87 (0.7–5.02) 
 IV 30/224 (13.4) 15/77 (19.5) 15/147 (10.2) 0.064 1,968 (1,322 ± 646) 23.22 (12.99–38.29) 11.34 (6.35–18.71) 0.05 2.02 (0.99–4.14) 0.037 2.13 (1.03–4.42) 
Mortality, n/total (%)Deaths per 1,000 person-years (95% CI)UnadjustedAdjusted
CategoryOverallrs966423-TTrs966423-CC/CTPPerson-years follow-uprs966423-TTrs966423-CC/CTLog-rank PHR (95%CI)Log-rank PHR (95% CI)
All patients 80/1,777 (4.5) 33/518 (6.4) 47/1,259 (3.7) 0.017 16,557 (11,709 ± 4,848) 6.81 (4.69–9.56) 4.01 (2.95–5.34) 0.022 1.67 (1.07–2.61) 0.038 1.6 (1.02–2.49) 
Gender 
 Female 66/1,586 (4.2) 26/454 (5.7) 40/1,132 (3.5) 0.052 14,913 (10,578 ± 4,335) 6 (3.92–8.79) 3.78 (2.7–5.15) 0.076 1.56 (0.95–2.55) 0.12 1.48 (0.9–2.42) 
 Male 14/191 (7.3) 7/64 (10.9) 7/127 (5.5) 0.238 1,644 (1,130 ± 514) 13.63 (5.48–28.08) 6.19 (2.49–12.76) 0.126 2.22 (0.78–6.35) 0.115 2.28 (0.79–6.57) 
Age_45 
 <45 7/741 (0.9) 2/214 (0.9) 5/527 (0.9) 7,325 (5,248 ± 2,077) 0.96 (0.12–3.48) 0.95 (0.31–2.22) 0.992 0.99 (0.19–5.11) 0.99 0.99 (0.19–5.1) 
 ≥45 73/1,036 (7) 31/304 (10.2) 42/732 (5.7) 0.016 9,232 (6,460 ± 2,772) 11.18 (7.6–15.88) 6.5 (4.69–8.79) 0.028 1.67 (1.05–2.67) 0.03 1.66 (1.04–2.65) 
Age_45_65 
 <45 7/741 (0.9) 2/214 (0.9) 5/527 (0.9) 7,325 (5,248 ± 2,077) 0.96 (0.12–3.48) 0.95 (0.31–2.22) 0.992 0.99 (0.19–5.11) 0.99 0.99 (0.19–5.1) 
 45–65 31/825 (3.8) 12/242 (5) 19/583 (3.3) 0.235 7,645 (5,348 ± 2,298) 5.22 (2.7–9.12) 3.55 (2.14–5.55) 0.356 1.4 (0.68–2.9) 0.357 1.41 (0.68–2.91) 
 >65 42/211 (19.9) 19/62 (30.6) 23/149 (15.4) 0.014 1,586 (1,112.5 ± 474) 40.08 (24.13–62.6) 20.67 (13.11–31.02) 0.035 1.91 (1.04–3.5) 0.038 1.89 (1.03–3.47) 
pT 
 pT1 28/1,085 (2.6) 9/298 (3) 19/787 (2.4) 0.668 9,514 (6,792 ± 2,722) 3.31 (1.51–6.28) 2.8 (1.68–4.37) 0.775 1.12 (0.51–2.49) 0.78 1.13 (0.49–2.59) 
 pT2 8/189 (4.2) 2/56 (3.6) 6/133 (4.5) 1,777 (1,278 ± 499) 4.01 (0.49–14.48) 4.69 (1.72–10.22) 0.824 0.83 (0.17–4.14) 0.576 0.63 (0.13–3.16) 
 pT3 29/362 (8) 17/118 (14.4) 12/244 (4.9) 0.003 3,265 (2,237.5 ± 1,028) 16.54 (9.63–26.48) 5.36 (2.77–9.37) 0.002 3.05 (1.46–6.4) 0.01 2.61 (1.23–5.53) 
 pT4 9/46 (19.6) 3/19 (15.8) 6/27 (22.2) 0.716 404 (241 ± 163) 18.38 (3.79–53.7) 24.9 (9.14–54.21) 0.733 0.79 (0.19–3.17) 0.775 0.81 (0.19–3.42) 
pN 
 N0 51/1,383 (3.7) 16/375 (4.3) 35/1,008 (3.5) 0.521 13,098 (9,440 ± 3,658) 4.37 (2.5–7.1) 3.71 (2.58–5.16) 0.649 1.15 (0.63–2.07) 0.781 1.09 (0.6–1.98) 
 N1a 6/171 (3.5) 4/65 (6.2) 2/106 (1.9) 0.202 1,492 (941 ± 551) 7.26 (1.98–18.6) 2.13 (0.26–7.68) 0.145 3.29 (0.6–17.97) 0.148 3.29 (0.6–18.1) 
 N1b 15/166 (9) 10/59 (16.9) 5/107 (4.7) 0.011 1,492 (987 ± 505) 19.8 (9.5–36.42) 5.07 (1.65–11.82) 0.008 3.84 (1.31–11.24) 0.018 3.47 (1.16–10.4) 
pM 
 M0 60/1,670 (3.6) 25/479 (5.2) 35/1,191 (2.9) 0.029 15,763 (11,172 ± 4,591) 5.45 (3.52–8.04) 3.13 (2.18–4.36) 0.04 1.7 (1.02–2.84) 0.065 1.62 (0.97–2.71) 
 M1 15/55 (27.3) 6/21 (28.6) 9/34 (26.5) 357 (226 ± 131) 45.86 (16.83–99.82) 39.78 (18.19–75.51) 0.775 1.16 (0.41–3.27) 0.198 1.97 (0.69–5.6) 
Capsule 
 N 27/1,024 (2.6) 19/750 (2.5) 8/274 (2.9) 0.826 9,780 (7,025.5 ± 2,754) 2.7 (1.63–4.22) 2.9 (1.25–5.72) 0.894 1.06 (0.46–2.42) 0.861 0.92 (0.38–2.22) 
 T0 20/347 (5.8) 12/234 (5.1) 8/113 (7.1) 0.469 2,989 (2,055 ± 934) 5.84 (3.02–10.2) 8.57 (3.7–16.88) 0.389 1.48 (0.6–3.62) 0.356 1.52 (0.62–3.72) 
 T1 25/312 (8) 11/211 (5.2) 14/101 (13.9) 0.013 2,739.5 (1,879 ± 861) 5.85 (2.92–10.48) 16.26 (8.89–27.29) 0.007 2.83 (1.28–6.24) 0.011 2.73 (1.22–6.11) 
Angioinvasion 
 N 38/1,253 (3) 12/362 (3.3) 26/891 (2.9) 0.718 11,939 (8,401 ± 3,538) 3.39 (1.75–5.92) 3.09 (2.02–4.53) 0.806 1.09 (0.55–2.16) 0.881 0.95 (0.47–1.93) 
 T 30/394 (7.6) 17/117 (14.5) 13/277 (4.7) 0.001 3,401 (2,441 ± 960) 17.71 (10.31–28.35) 5.33 (2.84–9.11) 0.001 3.28 (1.59–6.76) 3.48 (1.67–7.22) 
Multifocality 
 s 55/1,235 (4.5) 21/370 (5.7) 34/865 (3.9) 0.178 11,479 (8,065 ± 3,414) 6.15 (3.81–9.4) 4.22 (2.92–5.89) 0.157 1.48 (0.86–2.55) 0.161 1.48 (0.85–2.56) 
 m 17/459 (3.7) 8/125 (6.4) 9/334 (2.7) 0.091 3,921 (2,828 ± 1,093) 7.32 (3.16–14.42) 3.18 (1.46–6.04) 0.169 1.97 (0.73–5.3) 0.29 1.7 (0.63–4.61) 
Size, mm 
 0–9 15/606 (2.5) 6/153 (3.9) 9/453 (2) 0.226 5,052 (3,752 ± 1,300) 4.62 (1.69–10.05) 2.4 (1.1–4.55) 0.222 1.88 (0.67–5.29) 0.142 2.17 (0.75–6.26) 
 10–19 18/544 (3.3) 9/173 (5.2) 9/371 (2.4) 0.121 4,770 (3,187 ± 1,583) 5.69 (2.6–10.79) 2.82 (1.29–5.36) 0.235 1.77 (0.68–4.59) 0.216 1.81 (0.7–4.72) 
 20–29 7/212 (3.3) 0/59 (0) 7/153 (4.6) 0.194 2,016 (1,450 ± 566) 0 (0–6.52) 4.83 (1.94–9.94) 0.095 0 (0-Inf) 0.063 0 (0-Inf) 
 30–39 8/95 (8.4) 4/35 (11.4) 4/60 (6.7) 0.461 861 (571 ± 290) 13.78 (3.75–35.28) 7 (1.91–17.93) 0.304 2.04 (0.51–8.18) 0.342 1.94 (0.48–7.78) 
 >40 21/153 (13.7) 10/48 (20.8) 11/105 (10.5) 0.126 1,387 (991 ± 396) 25.26 (12.11–46.46) 11.1 (5.54–19.86) 0.064 2.2 (0.93–5.19) 0.2 1.75 (0.74–4.16) 
Stage 
 I 24/956 (2.5) 8/251 (3.2) 16/705 (2.3) 0.481 8,437 (6,120 ± 2,317) 3.45 (1.49–6.8) 2.61 (1.49–4.25) 0.632 1.23 (0.52–2.89) 0.722 1.18 (0.48–2.89) 
 II 6/151 (4) 1/45 (2.2) 5/106 (4.7) 0.67 1,465 (1,042 ± 423) 2.37 (0.06–13.18) 4.8 (1.56–11.2) 0.484 0.47 (0.06–4.05) 0.384 0.4 (0.05–3.42) 
 III 16/374 (4.3) 8/124 (6.5) 8/250 (3.2) 0.175 3,461 (2,346 ± 1,114) 7.18 (3.1–14.15) 3.41 (1.47–6.72) 0.137 2.07 (0.78–5.52) 0.208 1.87 (0.7–5.02) 
 IV 30/224 (13.4) 15/77 (19.5) 15/147 (10.2) 0.064 1,968 (1,322 ± 646) 23.22 (12.99–38.29) 11.34 (6.35–18.71) 0.05 2.02 (0.99–4.14) 0.037 2.13 (1.03–4.42) 

In patients with angioinvasion, the deaths per 1,000 person-years were 17.71 (95% CI, 10.31–28.35) versus 5.33 (95% CI, 2.84–9.11) in TT versus CC/CT patients (adjusted HR, 3.48; 95% CI, 1.67–7.22; log-rank P < 0.001]) as compared with 3.09 (95% CI, 2.02–4.53) in angioinvasion-negative CC/CT patients. In patients with extrathyroidal invasion, deaths per 1,000 person-years were 12.26 (95% CI, 7.68–18.56) versus 5.85 (95% CI, 3.71–8.77) in TT versus CC/CT patients (adjusted HR, 2.07 95% CI, 1.15–3.73; log-rank P = 0.013) as compared with 2.7 (95% CI, 1.63–4.22) in extrathyroidal invasion–negative CC/CT patients. In patients with lymph node metastasis, deaths per 1,000 person-years were 19.8 (95% CI, 9.5–36.42) versus 5.07 (95% CI, 1.65–11.82) in TT versus CC/CT patients (adjusted HR, 3.47; 95% CI, 1.16–10.4; log-rank P = 0.018) as compared with 3.71 (95% CI, 2.58–5.16) in lymph node metastasis–negative CC/CT patients. In patients with stage IV disease, the deaths per 1,000 person-years were 23.22 (95% CI, 12.99–38.29) versus 11.34 (95% CI, 6.35–18.71) in TT versus CC/CT patients (adjusted HR, 2.13; 95% CI, 1.03–4.42; log-rank P = 0.037) as compared with 3.05 (95% CI, 2.04–4.38) in stage I to III of CC/CT patients (Table 4, Fig. 2A).

In patients older than 45 years who carry the rs966423 risk genotype (TT), deaths per 1,000 person-years were 11.18 (95% CI, 7.6–15.88) as compared with 6.5 (95% CI, 4.69–8.79) in patients older than 45 years without the risk genotype (adjusted HR, 1.66; 95% CI, 1.04–2.65; log-rank P = 0.03; Table 4, Fig. 2B). In patients older than 65 years who carry the rs966423 risk genotype (TT), deaths per 1,000 person-years were 40.08 (95% CI, 24.13–62.6) as compared with 20.67 (95% CI, 13.11–31.02) in patients older than 65 years without the risk genotype (adjusted HR, 1.89; 95% CI, 1.03–3.47; logrank P = 0.038; Table 4, Fig. 2B).

Several germline changes have been reported to predispose to DTC, including variants in microRNA-146a, ATM-CHEK2-BRCA1 pathway, TERT gene, and 5 SNPs pinpointed by GWAS study (2–5, 12–15). So far, no study revealed the influence of germline (inherited) variants on the overall mortality of patients with thyroid cancer.

Analyzing the relationship between the 5 GWAS-derived variants and overall mortality of patients with thyroid cancer, we found the association of rs966423 with increased risk of death. As such, this study reveals the first genetic marker allowing for assessment of the clinical outcome of DTC prior to the surgery.

The rs966423 polymorphism was previously associated with thyroid-stimulating hormone levels and genetic predisposition to PTC in different populations (2, 4, 16), but its effect on the clinical outcome of cancer was unknown. The SNP is located within the DIRC3 gene found to be associated with renal cancer (17), but the data on its biologic role are still lacking. Although our previous study showed that the more common C allele predisposes to thyroid cancer (4), this study revealed that the rare T allele is associated with worse prognosis of patients with thyroid cancer. This is a common observation that different alleles of the same gene are associated with predisposition to cancer (initiation) or its clinical outcome (progression), as most recently shown by Zhao and colleagues (18).

In our cohort of 1,836 patients with DTC, the overall mortality amounted to 4.63% (85 of 1836) and, as expected on the basis of other clinical data, was significantly higher in patients with FTC (6.94%) and insular FTC (35%) compared with patients with PTC (3.9%). We replicated well-documented findings that extrathyroidal invasion, lymph node and distant metastases, as well as angioinvasion are risk factors leading to increased mortality of patients with cancer (19–21). Our results supported also the well-established impact of age on the outcome of thyroid cancer (22). In a study by Jonklaas and colleagues, the authors showed that the risk of death due to PTC was significantly lower in female than in male patients (HR, 0.40; 95% CI, 0.24–0.65, P < 0.001) but concluded that this advantage reflected a generalized gender survival advantage seen in the U.S. population (23). Moreover, after adjustment for the disease stage and age, women's outcomes were similar to men's. In our work, analyzing all conventional clinicopathologic risk factors in one multivariate model, we received comparable results. This analysis showed no survival advantage or disadvantage associated with sex; however, such factors as age older than 45 years, extrathyroidal invasion, and distant metastasis were all independently associated with worse prognosis. In the same analysis, we obtained positive results for the rs966423 variant, for which the overall mortality in TT-carriers was almost doubled (6.4% vs. 3.7%) compared with patients who were heterozygous or homozygous for the C allele (Fig. 1). In addition, the variant showed a significant additive interaction with several conventional risk factors including angioinvasion, lymph node metastasis, extrathyroidal invasion, and stage IV of the disease in tailoring the overall mortality of patients with thyroid cancer. For example, in patients with angioinvasion or distant lymph node metastasis, deaths per 1,000 person-years were 3.3- or 3.8-fold higher, respectively, in TT-carriers as than in CC/CT-carriers. Of note, all the HRs remained statistically significant also after adjustment for age and gender.

Interestingly, the analysis revealed that the overall mortality rates do not differ between patients with stage I, II, and III of thyroid cancer. It is only stage IV that significantly increases mortality by the mode of 4.3-fold compared with patients with lower stages of cancer. This observation is consistent with a number of other long-term observations indicating that staging is not an ideal indication of patients' survival (24).

The main strength of this study is a large number of analyzed patients who were treated in the same comprehensive cancer center, thus the overall outcome is free from the intercenter bias resulting from different treatment approaches. Also, all the samples were genotyped in a single experiment, thus the results are fully comparable. The presented results are therefore obtained in a coherent and fully controlled study. The most important finding of the study is the impact of rs966423 on the mortality of patients with thyroid cancer. Of note, although the C allele was associated with increased risk for PTC development, it is the T variant that doubles the overall mortality among patients with thyroid cancer.

The main limitation of the study is lack of validation cohort. Unfortunately, the study of the overall survival in cancers with low mortality requires especially large cohort with long enough observation time, and such cohort of patients with DTC is not available.

To our knowledge, the rs966423 variant is the only genetic risk marker whose clinical significance is still valid after adjustment for sex and age at diagnosis, as well as for lymph node metastasis, extrathyroidal invasion, angioinvasion, and distant metastasis in multivariant analysis. Contrary to BRAF mutation and other somatic changes putatively associated with patients' mortality, the status of germline rs966423 variant is known before the surgery and might be used in management of mortality risk by means of modification of therapy, nevertheless it requires further investigation in large prospective studies before it is ready for clinical application.

No potential conflicts of interest were disclosed.

Conception and design: M. Świerniak, A. Wójcicka, K. Jażdżewski

Development of methodology: M. Świerniak, M. Czetwertyńska, K. Jażdżewski

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Wójcicka, M. Czetwertyńska, J. Długosińska, E. Stachlewska, B. Górnicka, Ł. Koperski, M. Bogdańska, W. Wiechno, K. Jażdżewski

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Świerniak, A. Wójcicka, W. Gierlikowski, A. Kot, B. Górnicka, Ł. Koperski, M. Bogdańska, K. Jażdżewski

Writing, review, and/or revision of the manuscript: M. Świerniak, A. Wójcicka, M. Czetwertyńska, J. Długosińska, E. Stachlewska, W. Gierlikowski, A. Kot, B. Górnicka, Ł. Koperski, M. Bogdańska, W. Wiechno, K. Jażdżewski

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J. Długosińska, E. Stachlewska, W. Gierlikowski, A. Kot, W. Wiechno, K. Jażdżewski

Study supervision: K. Jażdżewski

The authors thank Dr. Jeff Palatini for critical reading of the article.

This work was supported by Polish National Science Centre Grant DEC-2013/11/B/NZ3/00193. The Laboratory of Genomic Medicine of the Medical University of Warsaw participates in BASTION, a program funded by the European Commission under the 7th Framework Program, and by the Ministry of Science and Higher Education (FP7-REGPOT-2012-CT2012-316254-BASTION). M. Świerniak, A. Wójcicka, M. Czetwertyńska, J Długosińska, and K. Jażdżewski were supported by the Foundation for Polish Science. W. Gierlikowski receives a fellowship from the Postgraduate School of Molecular Medicine.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Baldini
E
,
Sorrenti
S
,
Tuccilli
C
,
Prinzi
N
,
Coccaro
C
,
Catania
A
, et al
Emerging molecular markers for the prognosis of differentiated thyroid cancer patients
.
Int J Surg
2014
;
12
:
S52
6
.
2.
Gudmundsson
J
,
Sulem
P
,
Gudbjartsson
DF
,
Jonasson
JG
,
Masson
G
,
He
H
, et al
Discovery of common variants associated with low TSH levels and thyroid cancer risk
.
Nat Genet
2012
;
44
:
319
22
.
3.
Gudmundsson
J
,
Sulem
P
,
Gudbjartsson
DF
,
Jonasson
JG
,
Sigurdsson
A
,
Bergthorsson
JT
, et al
Common variants on 9q22.33 and 14q13.3 predispose to thyroid cancer in European populations
.
Nat Genet
2009
;
41
:
460
4
.
4.
Liyanarachchi
S
,
Wojcicka
A
,
Li
W
,
Czetwertynska
M
,
Stachlewska
E
,
Nagy
R
, et al
Cumulative risk impact of five genetic variants associated with papillary thyroid carcinoma
.
Thyroid
2013
;
23
:
1532
40
.
5.
Jazdzewski
K
,
Murray
EL
,
Franssila
K
,
Jarzab
B
,
Schoenberg
DR
,
de la Chapelle
A
. 
Common SNP in pre-miR-146a decreases mature miR expression and predisposes to papillary thyroid carcinoma
.
Proc Natl Acad Sci U S A
2008
;
105
:
7269
74
.
6.
Landa
I
,
Robledo
M
. 
Association studies in thyroid cancer susceptibility: are we on the right track
?
J Mol Endocrinol
2011
;
47
:
R43
58
.
7.
Liu
X
,
Bishop
J
,
Shan
Y
,
Pai
S
,
Liu
D
,
Murugan
AK
, et al
Highly prevalent TERT promoter mutations in aggressive thyroid cancers
.
Endocr Relat Cancer
2013
;
20
:
603
10
.
8.
Melo
M
,
da Rocha
AG
,
Vinagre
J
,
Batista
R
,
Peixoto
J
,
Tavares
C
, et al
TERT promoter mutations are a major indicator of poor outcome in differentiated thyroid carcinomas
.
J Clin Endocrinol Metab
2014
;
99
:
E754
65
.
9.
Xing
M
,
Alzahrani
AS
,
Carson
KA
,
Viola
D
,
Elisei
R
,
Bendlova
B
, et al
Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer
.
JAMA
2013
;
309
:
1493
501
.
10.
Ulisse
S
,
Baldini
E
,
Sorrenti
S
,
Barollo
S
,
Prinzi
N
,
Catania
A
, et al
In papillary thyroid carcinoma BRAFV600E is associated with increased expression of the urokinase plasminogen activator and its cognate receptor, but not with disease-free interval
.
Clin Endocrinol (Oxf)
2012
;
77
:
780
6
.
11.
Andersen
P
,
Gill
R
. 
Cox's regression model for counting processes, a large sample study
.
Ann Stat
1982
;
10
:
1100
120
.
12.
Wojcicka
A
,
Czetwertynska
M
,
Swierniak
M
,
Długosińska
J
,
Maciąg
M
,
Czajka
A
, et al
Variants in the ATM-CHEK2-BRCA1 axis determine genetic predisposition and clinical presentation of papillary thyroid carcinoma
.
Genes Chromosomes Cancer
2014
;
53
:
516
23
.
13.
Jazdzewski
K
,
Liyanarachchi
S
,
Swierniak
M
,
Pachucki
J
,
Ringel
MD
,
Jarzab
B
, et al
Polymorphic mature microRNAs from passenger strand of pre-miR-146a contribute to thyroid cancer
.
Proc Natl Acad Sci U S A
2009
;
106
:
1502
05
.
14.
Cantara
S
,
Pisu
M
,
Frau
DV
,
Caria
P
,
Dettori
T
,
Capezzone
M
, et al
Telomere abnormalities and chromosome fragility in patients affected by familial papillary thyroid cancer
.
J Clin Endocrinol Metab
2012
;
97
:
E1327
31
.
15.
Cybulski
C
,
Gorski
B
,
Huzarski
T
,
Masojć
B
,
Mierzejewski
M
,
Dębniak
T
, et al
CHEK2 is a multiorgan cancer susceptibility gene
.
Am J Hum Genet
2004
;
75
:
1131
5
.
16.
Wang
YL
,
Feng
SH
,
Guo
SC
,
Wei
WJ
,
Li
DS
,
Wang
Y
, et al
Confirmation of papillary thyroid cancer susceptibility loci identified by genome-wide association studies of chromosomes 14q13, 9q22, 2q35 and 8p12 in a Chinese population
.
J Med Genet
2013
;
50
:
689
95
.
17.
Bodmer
D
,
Schepens
M
,
Eleveld
MJ
,
Schoenmakers
EF
,
Geurts van Kessel
A
. 
Disruption of a novel gene, DIRC3, and expression of DIRC3-HSPBAP1 fusion transcripts in a case of familial renal cell cancer and t(2;3)(q35;q21)
.
Genes Chromosomes Cancer
2003
;
38
:
107
16
.
18.
Zhao
L
,
Zhang
Q
,
Luan
X
,
Huang
X
,
Zhao
S
,
Zhao
H
. 
STAT3 and STAT5b polymorphism contributes to breast cancer risk and clinical outcomes
.
Int J Clin Exp Pathol
2015
;
8
:
2033
8
.
19.
Kim
TY
,
Hong
SJ
,
Kim
JM
,
Kim
WG
,
Gong
G
,
Ryu
JS
, et al
Prognostic parameters for recurrence of papillary thyroid microcarcinoma
.
BMC Cancer
2008
;
8
:
296
.
20.
Kilfoy
BA
,
Devesa
SS
,
Ward
MH
,
Zhang
Y
,
Rosenberg
PS
,
Holford
TR
, et al
Gender is an age-specific effect modifier for papillary cancers of the thyroid gland
.
Cancer Epidemiol Biomarkers Prev
2009
;
18
:
1092
100
.
21.
Johnston
LE
,
Tran Cao
HS
,
Chang
DC
,
Bouvet
M
. 
Sociodemographic Predictors of Survival in Differentiated Thyroid Cancer: Results from the SEER Database
.
ISRN Endocrinol
2012
;
2012
:
384707
.
22.
Mazzaferri
EL
,
Jhiang
SM
. 
Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer
.
Am J Med
1994
;
97
:
418
28
.
23.
Jonklaas
J
,
Nogueras-Gonzalez
G
,
Munsell
M
,
Litofsky
D
,
Ain
KB
,
Bigos
ST
, et al
The impact of age and gender on papillary thyroid cancer survival
.
J Clin Endocrinol Metab
2012
;
97
:
E878
7
.
24.
Edge
SB
,
Compton
CC
. 
The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM
.
Ann Surg Oncol
2010
;
17
:
1471
4
.