Prognostic factors in patients with thyroid carcinoma: a competing-risks analysis

doi:10.21203/rs.3.rs-19798/v1

Background

Cox proportional-hazards models are widely used to describe survival trends and identify prognostic factors for thyroid carcinoma, but they have significant limitations and deficiencies. This study therefore used a competing-risks model to identify the significant prognostic factors for thyroid carcinoma.

Methods

We identified 38,444 eligible patients in the SEER (Surveillance, Epidemiology, and End Result) database. The potential prognostic factors for thyroid carcinoma were analyzed by competing-risks analysis using both univariate and multivariate analyses.

Results

The univariate analysis showed that age, sex, race, marital state, insurance status, tumor size, whether regional lymph nodes were examined, AJCC stage, histology, surgery status, radiation status, chemotherapy status, bone metastasis, brain metastasis, liver metastasis, and lung metastasis were prognostic factors for death caused by thyroid carcinoma. The multivariate analyses that comprised Cox regression analysis, the cause-specific hazard function analysis, and subdistribution hazard function (SD) analysis produced different results, identifying age, being unmarried, no regional lymph nodes examined, AJCC stages II, III, and IV, having follicular, medullary, and anaplastic carcinomas, no surgery, no radiation, liver metastasis, and lung metastasis as the significant risk factors for thyroid carcinoma, while being female and not receiving chemotherapy were protective factors. The results from the three multivariate models for being black, tumor size >1 cm, and brain metastasis were inconsistent.

Conclusion

This study had produced information about the significant prognostic factors for thyroid carcinoma using a competing-risks model that is more accurate than that obtained using Cox regression analysis. The SD model seems to be preferable for establishing a more accurate prognostic model of this disease aimed at guiding clinical treatments and improving prognoses.

Endocrinology & Metabolism

competing-risks analysis

SEER

thyroid carcinoma

The detection rate of thyroid cancer is continuing to increase. Its incidence is increasing faster than that of any other solid tumor, and it was ranked in ninth place for incidence in 2018 and is predicted to replace colorectal cancer as the fourth leading cancer diagnosis by 2030 in the USA. Although thyroid cancer has an excellent prognosis, it still kills estimated 41,000 people annually worldwide(1). The main pathological types of thyroid carcinoma are papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC), anaplastic thyroid carcinoma (ATC), and medullary thyroid carcinoma (MTC)(2), with PTC constituting 90% of them. Most cases have an excellent prognosis, but approximately 10% of PTC patients exhibit recurrences such as lung metastasis and lymph node recurrence(3), or even death. Although much less common, FTC is widely invasive, with a reported mortality rate of around 20%(4). ATC is an extremely aggressive undifferentiated tumor, with almost 100% disease-specific mortality(5), and while representing about only < 2% of thyroid cancers ATC is responsible for 40% of thyroid cancer deaths. MTC represents < 5% of thyroid carcinomas, and the overall 5-year survival rate of patients with MTC is 86%(6).

Adequate risk stratification is crucial in a malignant neoplastic disease in order to avoid both the undertreatment of high-risk subjects and the overtreatment of low-risk patients(7). This means that personalized treatment according to the potential prognosis of individual patients with thyroid cancer is critically important. Cox regression is currently the most widely method used to predict the risk factors for thyroid cancer(8–11). However, clinical survival data are often associated with multiple outcomes, during which there may be competing relationships, and the use of Cox regression analysis may lead to inaccurate estimates of hazard ratios (HRs)(12, 13). A competing-risks model is better than a Cox regression model in such situations in clinical epidemiology(14–16), but competing-risks analyses of the prognostic factors in patients with thyroid carcinoma have rarely been reported. The present study analyzed two types of events: death due to thyroid carcinoma and death due to other causes (DOC). A competing-risks model was adopted in this study to identify the significant prognostic factors for thyroid carcinoma in order to guide clinical treatment more accurately.

A competing risk refers to a situation where an individual is exposed to multiple causes of failure, with the eventual failure being attributed to exactly one of them. In this situation, the occurrence of one type of event hinders the occurrence of any other event(17). The failure events in the present study were death due to thyroid carcinoma and DOC, with the latter including car collisions, stroke, and suicide. We used the competing-risks model of SAS statistical software to analyze the survival of patients with thyroid carcinoma in order to obtain more accurate information on the significant prognostic factors for death due to thyroid carcinoma.

Patients

The patient data were obtained from the latest version of the SEER (Surveillance, Epidemiology, and End Result) database using SEER*Stat software (version 8.3.5 released on April 15, 2019; https://seer.cancer.gov/). This study was approved by the Ethics Committee of the Ninth Hospital of Xi’an. The following inclusion criteria were applied:

Presence of thyroid carcinoma according to ICD-O-3/WHO 2008 histological type codes of 8330/3 (FTC), 8050/3 (PTC), 8510/3 (MTC), and 8021/3 (ATC).
Positive histology diagnostic confirmation.
Categorized as alive, death due to thyroid cancer, or DOC.
Active follow-up.
Age at diagnosis of 19–85 years.

The exclusion criteria were as follows:

Unknown age, race, sex, or AJCC stage, or incomplete SEER cause-specific death classification.
Unknown survival time for a patient who was still alive.
Diagnosis based only a death certificate or an autopsy.

We collected the following data for each patient: age, race, sex, marital status, insurance status, tumor size, whether regional lymph nodes were examined, AJCC stage, histology, surgery status, radiation status, chemotherapy status, bone metastases, brain metastases, liver metastases, lung metastases, survival time (in months), and case outcome. The case outcome was divided into the following three conditions: alive, death due to thyroid carcinoma (endpoint event), and DOC (competing event).

The application of the inclusion and exclusion criteria resulted in the identification of 38,444 patients in the SEER database between 2010 and 2015.

Statistical analysis

All statistical analyses were performed using SPSS (version 21.0) and SAS statistical software. Mean±standard-deviation values were used to express continuous variables conforming to a normal distribution, and all other variables were expressed as median (25th–75th percentile) values. The cumulative incidence function was used in the univariate analysis to analyze each potential prognostic factor, and the value obtained in Gray’s test and the cumulative incidence rate at each time point were obtained. Cox regression analysis, cause-specific hazard function (CS) analysis, and subdistribution hazard function (SD) analysis were used for the multivariate analyses, and independent prognostic factors were obtained. Probability values of p<0.05 were considered statistically significant.

Patient characteristics

The median age of all of the patients was 46 years, with an age range from 35 to 57 years. The median age of patients who died due to thyroid cancer was 68 years, with an age range from 57 to 77 years. Most of the patients in the total patient group and in the group of those who died due to thyroid cancer were female, white, and married, had insurance and any medical, a tumor size of £1 cm, and PTC. Most of the patients in both groups had had their regional lymph nodes examined, had received surgery, and had received chemotherapy. Most of the patients who died due to thyroid cancer were in AJCC stage IC, had not received radiation, and had an unknown metastasis status. Most of the patients in the total patient group were in AJCC stage I, had received radiation, and did not have metastasis. The demographics and tumor characteristics of the patients are summarized in Table 1.

Univariate analysis

Gray’s test was used to perform a univariate analysis of the following 16 potential prognostic factors: age, sex, race, marital state, insurance status, tumor size, whether regional lymph nodes were examined, AJCC stage, histology, surgery status, radiation status, chemotherapy status, bone metastasis, brain metastasis, liver metastasis, and lung metastasis. All 16 factors were found to significantly affect the prognosis for death caused by thyroid carcinoma (p<0.05). In addition, the cumulative incidence rates at 50 and 100 months were calculated, as presented in Figure 1 and Table 2.

Multivariate analyses

The application of Cox regression analysis, CS analysis, and SD analysis for the multivariate analyses produced different results. Age, being unmarried, no regional lymph nodes examined, AJCC stages II, III, and IV, having FTC, MTC, and ATC, no surgery, no radiation, liver metastasis, and lung metastasis were found to be significant risk factors for thyroid carcinoma in all three methods, while being female and not receiving chemotherapy were protective factors. The HR values for these predictive factors differed between the three models.

Cox regression analysis showed that being black (p=0.01, HR=1.25) was a risk factor, while the SD analysis (p=0.6) and CS analysis (p=0.25) did not. Cox regression analysis showed that tumor size >1 cm (p=0.13) was not statistically significant, while the SD analysis (p<0.01, HR=2.48) and CS analysis (p=0.01, HR=2.59) showed that this was a significant risk factor. Cox regression analysis (p<0.01, HR=1.93) and CS analysis (p=0.01, HR=1.77) indicated that brain metastasis was a risk factor, while the SD analysis (p=0.09) did not (Table 3).

The sample size should be considered carefully when designing a competing-risks analysis. If the proportion of competing events is greater than 10%, a Cox regression analysis can be severely affected by bias(18) that overestimates the event incidence and poorly estimates the HRs. The present study included 1205 DOC patients, representing almost half of the deaths, and so we used a competing-risks model that could avoid this bias to obtain more accurate prognostic factors for thyroid carcinoma. Competing-risks regression approaches focus on two definitions of hazard: SD and CS. The SD model is useful for predicting an individual’s risk or when allocating resources, while the CS model may be better suited for studying the etiology of diseases(19). Since these two models have their own unique underlying mechanisms, it is necessary to provide the results obtained from SD and CS models simultaneously. Koller et al.(20) proposed that a SD model tends to estimate the disease risk and prognosis, which is more suitable for establishing a clinical prediction model and risk scores. It is obvious that the HRs obtained in the present SD model were the most valuable, since this model focuses on the direct assessment of actual risks and therefore also the prognosis and medical decision-making.

The Cox regression analysis showed that there was no significant difference between tumor sizes of > 1 and ≤1 cm, and that being black was a risk factor in the thyroid carcinoma patients. However, the competing-risks model produced the opposite results. There is a considerable amount of evidence that tumor size is an important prognostic factor for thyroid carcinoma(21). Meanwhile, the effect of race in predicting thyroid cancer death was inconclusive in our competing-risks model. In could be that the Cox analysis overestimated the HR value for race, resulting in a false-positive result.

All three methods applied in this study indicated that older age, FTC, MTC, and ATC (all vs PTC), AJCC stages II, III, and IV (all vs stage I), liver metastasis, and lung metastasis were detrimental prognostic factors. It is well known that age > 45 years, extrathyroidal invasion, distant metastasis, large tumor, vascular invasion, and poor differentiated histology are detrimental prognostic factors(4, 22, 23).

We found that the female-specific mortality was lower than the male-specific mortality in all three methods, even though the incidence rate is higher in females than males(24). There are conflicting reports on the effect of sex on mortality. Ito and Miyauchi(25) and Lin et al.(26) found that a larger proportion of male than female patients died of thyroid cancer, whereas Nguyen et al.(27) found the opposite result. Bray et al.(24) reported that the mortality rates were similar in the sexes, based on global cancer statistics published in 2018. It is possible that any effect of sex on mortality is related to geography and race. More evidence is needed on the effect of sex on thyroid-cancer-specific death.

All three methods utilized in our study indicated that having the regional lymph nodes removed and examined by the pathologist was a significant protective factor for thyroid-cancer-specific death. There is no report in the literature on whether having regional lymph nodes examined affects the prognosis of thyroid cancer, while there is substantial evidence that local lymph node metastasis is a significant risk factor for thyroid-cancer-related death(28–32). This means that removing and examining the lymph nodes during surgery can improve the prognosis. As described in Shi et al.(33) and Merrill and Johnson(34), we also found that being married has a positive effect on the prognosis of thyroid cancer in all three analysis methods.

The standard treatment for thyroid cancer is surgery followed by radioactive iodine or observation. Chemotherapy is only considered if the patient has any of the following conditions: (1) clinically significant disease and evidence of disease progression, (2) a symptomatic tumor burden that cannot be managed with localized treatments or other medical treatment, or (3) the tumor threatens vital structures and cannot be managed with localized treatments(35). Still, the effects of chemotherapy have not been proven(36–39). We found that receiving surgery and radiation therapy were protective factors, while chemotherapy was a risk factor for this disease.

While the above-mentioned factors have been demonstrated to be independent factors for the prognosis of thyroid cancer in three analysis methods, compared to the SD model, Cox regression overrated the prognostic effect of almost all of the variables we investigated, including age, sex, marital state, surgery status, radiation status, and metastasis, as also described by Pintilie(18).

This study has provided information about the significant prognostic factors for thyroid carcinoma using a competing-risks model that is more accurate than that obtained in Cox regression analysis. The SD model seems to provide a more accurate prognostic model of this disease aimed at guiding clinical treatments and improving prognoses.

Authors' contributions

(1) Conception and design: Junhu Wang and Lisong Heng. (2) Administrative support: Xinwen Wang, Junhu Wang. (3) Provision of study materials or patients: Jie Yang and Xinwen Wang.(4) Collection and assembly of data: Lisong Heng, Xinwen Wang and Feng Tian. (5) Data analysis and interpretation: Xiaojun Liang, Xinwen Wang, Lisong Heng and Junhu Wang. (6) Manuscript writing: Junhu Wang and Lisong Heng. (7) Final approval of manuscript: Junhu Wang, Lisong Heng, Jie Yang, Feng Tian, Xiaojun Liang and Xinwen Wang.

Acknowledgment

We thank all colleagues involved in the study for their contributions.

Funding: None

Competing interests

The author reports have no conflicts of interest in this work

Notes

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Ninth Hospital of Xi’an

Consent for publication

All patients came from the SEER database (Surveillance, Epidemiology, and End Result), which is publicly available.

Availability of data and material

The datasets analyzed during current study are available from the corresponding author upon reasonable request.

Rahib L., Smith B. D., Aizenberg R., Rosenzweig A. B., Fleshman J. M., Matrisian L. M. Projecting Cancer Incidence and Deaths to 2030: The Unexpected Burden of Thyroid, Liver, and Pancreas Cancers in the United States. CANCER RES 2014;74: 2913-21.
Jemal A., Bray F., Center M. M., Ferlay J., Ward E., Forman D. Global cancer statistics. CA Cancer J Clin 2011;61: 69-90.
Mansour J., Sagiv D., Alon E., Talmi Y. Prognostic value of lymph node ratio in metastatic papillary thyroid carcinoma. J LARYNGOL OTOL 2018;132: 8-13.
Vuong H. G., Kondo T., Pham T. Q., Oishi N., Mochizuki K., Nakazawa T., et al.. Prognostic significance of diffuse sclerosing variant papillary thyroid carcinoma: a systematic review and meta-analysis. EUR J ENDOCRINOL 2017;176: 431-9.
Are C., Shaha A. R. Anaplastic thyroid carcinoma: biology, pathogenesis, prognostic factors, and treatment approaches. ANN SURG ONCOL 2006;13: 453-64.
Giuffrida D., Gharib H. Current diagnosis and management of medullary thyroid carcinoma. ANN ONCOL 1998;9: 695-701.
Krajewska Jolanta, Chmielik Ewa, Jarząb Barbara. Dynamic risk stratification in the follow-up of thyroid cancer: what is still to be discovered in 2017? ENDOCR-RELAT CANCER 2017;24: R387-402.
Guan Y. J., Fang S. Y., Chen L. L., Li Z. D. Development and validation of prognostic nomograms for medullary thyroid cancer. Onco Targets Ther 2019;12: 2299-309.
Zhang N., Cong X., Zhou D., Guo L., Yuan C., Xu D., et al.. Predictive significance of serum dipeptidyl peptidase-IV in papillary thyroid carcinoma. CANCER BIOMARK 2019;24: 7-17.
Nam S. H., Bae M. R., Roh J. L., Gong G., Cho K. J., Choi S. H., et al.. A comparison of the 7th and 8th editions of the AJCC staging system in terms of predicting recurrence and survival in patients with papillary thyroid carcinoma. ORAL ONCOL 2018;87: 158-64.
Teng H., Mao F., Liang J., Xue M., Wei W., Li X., et al.. Transcriptomic signature associated with carcinogenesis and aggressiveness of papillary thyroid carcinoma. THERANOSTICS 2018;8: 4345-58.
Putter H., Fiocco M., Geskus R. B. Tutorial in biostatistics: competing risks and multi-state models. STAT MED 2007;26: 2389-430.
Berry S. D., Ngo L., Samelson E. J., Kiel D. P. Competing risk of death: an important consideration in studies of older adults. J AM GERIATR SOC 2010;58: 783-7.
Andersen P. K., Pohar Perme M. Inference for outcome probabilities in multi-state models. LIFETIME DATA ANAL 2008;14: 405-31.
Austin P. C., Lee D. S., Fine J. P. Introduction to the Analysis of Survival Data in the Presence of Competing Risks. CIRCULATION 2016;133: 601-9.
Lau B., Cole S. R., Gange S. J. Competing risk regression models for epidemiologic data. AM J EPIDEMIOL 2009;170: 244-56.
Austin P. C., Lee D. S., Fine J. P. Introduction to the Analysis of Survival Data in the Presence of Competing Risks. CIRCULATION 2016;133: 601-9.
Pintilie M. Dealing with competing risks: testing covariates and calculating sample size. STAT MED 2002;21: 3317-24.
Lau B., Cole S. R., Gange S. J. Competing risk regression models for epidemiologic data. AM J EPIDEMIOL 2009;170: 244-56.
Koller M. T., Raatz H., Steyerberg E. W., Wolbers M. Competing risks and the clinical community: irrelevance or ignorance? STAT MED 2012;31: 1089-97.
Perrier N. D., Brierley J. D., Tuttle R. M. Differentiated and anaplastic thyroid carcinoma: Major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin 2018;68: 55-63.
Ito Y., Hirokawa M., Higashiyama T., Takamura Y., Miya A., Kobayashi K., et al.. Prognosis and prognostic factors of follicular carcinoma in Japan: importance of postoperative pathological examination. WORLD J SURG 2007;31: 1417-24.
Elisei R., Cosci B., Romei C., Bottici V., Renzini G., Molinaro E., et al.. Prognostic significance of somatic RET oncogene mutations in sporadic medullary thyroid cancer: a 10-year follow-up study. J Clin Endocrinol Metab 2008;93: 682-7.
Bray F., Ferlay J., Soerjomataram I., Siegel R. L., Torre L. A., Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68: 394-424.
Ito Y., Miyauchi A. Prognostic factors and therapeutic strategies for differentiated carcinomas of the thyroid. ENDOCR J 2009;56: 177-92.
Lin J. D., Hsueh C., Chao T. C. Long-Term Follow-Up of the Therapeutic Outcomes for Papillary Thyroid Carcinoma With Distant Metastasis. Medicine (Baltimore) 2015;94: e1063.
Nguyen M. T., Hu J., Hastings K. G., Daza E. J., Cullen M. R., Orloff L. A., et al.. Thyroid cancer mortality is higher in Filipinos in the United States: An analysis using national mortality records from 2003 through 2012. CANCER-AM CANCER SOC 2017;123: 4860-7.
Yamashita H., Noguchi S., Murakami N., Kawamoto H., Watanabe S. Extracapsular invasion of lymph node metastasis is an indicator of distant metastasis and poor prognosis in patients with thyroid papillary carcinoma. CANCER-AM CANCER SOC 1997;80: 2268-72.
Yamashita H., Noguchi S., Murakami N., Toda M., Uchino S., Watanabe S., et al.. Extracapsular invasion of lymph node metastasis. A good indicator of disease recurrence and poor prognosis in patients with thyroid microcarcinoma. CANCER-AM CANCER SOC 1999;86: 842-9.
Ito Y., Fukushima M., Tomoda C., Inoue H., Kihara M., Higashiyama T., et al.. Prognosis of patients with papillary thyroid carcinoma having clinically apparent metastasis to the lateral compartment. ENDOCR J 2009;56: 759-66.
Ricarte-Filho J., Ganly I., Rivera M., Katabi N., Fu W., Shaha A., et al.. Papillary thyroid carcinomas with cervical lymph node metastases can be stratified into clinically relevant prognostic categories using oncogenic BRAF, the number of nodal metastases, and extra-nodal extension. THYROID 2012;22: 575-84.
Wu M. H., Shen W. T., Gosnell J., Duh Q. Y. Prognostic significance of extranodal extension of regional lymph node metastasis in papillary thyroid cancer. Head Neck 2015;37: 1336-43.
Shi R. L., Qu N., Lu Z. W., Liao T., Gao Y., Ji Q. H. The impact of marital status at diagnosis on cancer survival in patients with differentiated thyroid cancer. Cancer Med 2016;5: 2145-54.
Merrill R. M., Johnson E. Benefits of marriage on relative and conditional relative cancer survival differ between males and females in the USA. J CANCER SURVIV 2017;11: 578-89.
Cabanillas M. E., McFadden D. G., Durante C. Thyroid cancer. LANCET 2016;388: 2783-95.
Brose M. S., Nutting C. M., Jarzab B., Elisei R., Siena S., Bastholt L., et al.. Sorafenib in radioactive iodine-refractory, locally advanced or metastatic differentiated thyroid cancer: a randomised, double-blind, phase 3 trial. LANCET 2014;384: 319-28.
Schlumberger M., Tahara M., Wirth L. J., Robinson B., Brose M. S., Elisei R., et al.. Lenvatinib versus placebo in radioiodine-refractory thyroid cancer. N Engl J Med 2015;372: 621-30.
Wells SA Jr, Robinson B. G., Gagel R. F., Dralle H., Fagin J. A., Santoro M., et al.. Vandetanib in patients with locally advanced or metastatic medullary thyroid cancer: a randomized, double-blind phase III trial. J CLIN ONCOL 2012;30: 134-41.
Elisei R., Schlumberger M. J., Muller S. P., Schoffski P., Brose M. S., Shah M. H., et al.. Cabozantinib in progressive medullary thyroid cancer. J CLIN ONCOL 2013;31: 3639-46.

Table 1 Clinical and pathological characteristics of patients with thyroid cancer.

Prognostic factors	Classification	All	Death due to thyroid cancer
Number of Patients n		38444	1210
Median age at diagnosis, year (interquartile range)		46(35-57)	68（57.75-77）
Sex n (%)
	Male	9095(23.6)	537（44.38）
	Female	29349(76.3)	673（55.62）
Race n (%)
	White	30589(79.57)	916（75.70）
	Black	1912(4.97)	78（6.45）
	Other	5943(15.46)	216（17.85）
Married State n (%)
	Married	23508（61.14）	662（54.71）
	Unmarried	13209（34.36）	507（41.90）
	Unknown	1727（4.49）	41（3.38）
Insurance Recode n (%)
	Uninsurance	719（1.87）	25（2.07）
	Insurance and Any Medical	30135（78.38）	832（68.76）
	Unknown	7590（19.74）	353（29.17）
Tumor Size n (%)
	≤1cm	37567（97.72）	969（80.08）
	>1cm	109（0.28）	7（0.58）
	Unknown	768（2.00）	234（19.34）
Regional nodes Examined n (%)
	Yes	22695（59.03）	611（50.50）
	None	15672（40.77）	585（48.35）
	Unknown	77（0.20）	14（1.16）
AJCC stage n (%)
	I	27321（71.07）	61（5.04）
	II	2380（6.19）	40（3.31）
	III	4895（12.73）	99（8.18）
	IV	3848（10.01）	1010（83.47）
histology n (%)
	Papillary	34624（90.06）	592（48.93）
	Follicular	2411（6.27）	141（11.65）
	Medullary	981（2.55）	103（8.51）
	anaplastic	428（1.11）	374（30.91）

Prognostic factors	Classification	All	Death due to thyroid cancer
Surgery
	Yes	37688（98.03）	826（68.26）
	No/Unknown	756（1.97）	384（31.74）
Radiation
	Yes	20483（53.28）	575（47.52）
	None	17961（46.72）	635（52.48）
Chemotherapy
	Yes	383（1.00）	268（22.15）
	No/Unknown	38061（99.0）	942（77.85）
Mets at bone
	No	20972（54.55）	425（35.12）
	Yes	145（0.38）	68（5.62）
	Unknown	17327（45.07）	717（59.26）
Mets at brain
	No	20972（54.55）	464（38.35）
	Yes	145（0.38）	21（1.74）
	Unknown	17327（45.07）	725（59.92）
Mets at liver
	No	21068（54.80）	463（38.26）
	Yes	40（0.10）	24（1.98）
	Unknown	17336（45.09）	723（59.75）
Mets at lung
	No	20816（54.15）	321（26.53）
	Yes	302（0.79）	174（14.38）
	Unknown	17326（45.07）	715（59.09）

Regional nodes Examined: whether regional lymph nodes had been removed and examined by the pathologist.

Table 2 Univariate analysis of prognostic factors in patients with thyroid cancer.

Prognostic factors	Classification	Gray's test	P-value	50-month CIF	100-month CIF
Age		4078.66	<0.0001
Sex		294.206	<0.0001
	Male			0.049501	0.071436
	Female			0.019922	0.026577
Race		14.8218	0.0006
	White			0.025628	0.034991
	Black			0.036149	0.048238
	Other			0.030713	0.045734
Married State		34.5785	<.0001
	Married			0.023477	0.032510
	Unmarried			0.034271	0.045952
	Unknown			0.017915	0.034532
Insurance Recode		8.87786	0.0118
	Uninsurance			0.034382
	Insurance and Any Medical			0.025231	0.035461
	Unknown			0.032116	0.041782
Tumor Size		1864.36	<.0001
	≤1cm			0.02148	0.03088
	>1cm			0.06301
	Unknown			0.29196	0.33284
Regional nodes Examined		85.9680	<.0001
	Yes			0.02241	0.03430
	None			0.03295	0.04086
	Unknown			0.16628
AJCC stage		7623.33	<.0001
	I			0.00139	0.00319
	II			0.01093	0.02153
	III			0.01197	0.02727
	IV			0.23749	0.30195
histology		11641.30	<0.0001
	Papillary			0.01330	0.02145
	Follicular			0.04396	0.06957
	Medullary			0.09138	0.12627
	anaplastic			0.88337

Prognostic factors	Classification	Gray's test	P-value	50-month CIF	100-month CIF
Surgery		6649.63	<0.0001
	Yes			0.01732	0.02705
	No/Unknown			0.52500	0.56066
Radiation		26.5685	<0.0001
	Yes			0.021760	0.033673
	None			0.032753	0.040686
Chemotherapy		5824.72	<.0001
	No/Unknown			0.69220	0.74011
	Yes			0.02022	0.02992
Mets at bone		1214.47	<.0001
	No			0.02094
	Yes			0.48200
	Unknown			0.02983	0.03965
Mets at brain		795.594	<.0001
	No			0.02274	0.6359
	Yes
	Unknown			0.03028	0.04009
Mets at liver		589.715	<.0001
	No			0.02276
	Yes			0.68281
	Unknown			0.03017	0.03998
Mets at lung		3789.35	<.0001
	No			0.01591
	Yes			0.59307
	Unknown			0.02972	0.03954

Race-Other：American Indian & AK Native & Asian & Pacific Islander

Marital status-Unmarried：Single & Separated & Divorced &Widowed &Unmarried or Domestic Partner

Table 3 Multivariate analysis of prognostic factors in patients with thyroid cancer

Classification	Cox regression analysis			SD model analysis			CS model analysis
Classification	P-value	HR	95%CI	P-value	HR	95%CI	P-value	HR	95%CI
Age	<0.01	1.06	1.06-1.07	<0.01	1.04	1.03-1.04	<0.01	1.05	1.04-1.05
Sex
Male	-	-	-	-	-	-	-	-	-
Female	<0.01	0.63	0.58-0.69	0.02	0.84	0.72-0.97	<0.01	0.78	0.69-0.87
Race
White	-	-	-	-	-	-	-	-	-
Black	0.01	1.25	1.06-1.47	0.60	1.079	0.82-1.43	0.25	1.16	0.91-1.47
Other	0.09	0.91	0.81-1.02	0.35	1.08	0.92-1.26	0.79	1.02	0.88-1.19
Married State
Married	-	-	-	-	-	-	-	-	-
Unmarried	<0.01	1.64	1.50-1.79	<0.01	1.328	1.14-1.55	<0.01	1.43	1.26-1.62
Unknown	0.28	1.12	0.91-1.38	0.58	0.91	0.65-1.28	0.56	0.91	0.66-1.25
Insurance Recode
Uninsurance	-	-	-	-	-	-	-	-	-
Insurance and Any Medical	0.50	0.89	0.62-1.26	0.07	0.67	0.43-1.04	0.13	0.73	0.48-1.10
Unknown	0.80	1.05	0.73-1.50	0.59	0.88	0.56-1.40	0.61	0.90	0.58-1.38
Tumor Size
≤1cm	-	-	-	-	-	-	-	-	-
>1cm	0.13	1.59	0.88-2.88	<0.01	2.48	1.37-4.49	0.01	2.59	1.23-5.47
Unknown	<0.01	1.50	1.30-1.74	<0.01	1.52	1.20-1.92	<0.01	1.71	1.45-2.02
Regional nodes Examined
Yes	-	-	-	-	-	-	-	-	-
None	<0.01	1.25	1.14-1.37	0.01	1.25	1.06-1.47	<0.01	1.31	1.15-1.51
Unknown	0.37	1.26	0.76-2.09	0.01	1.81	1.14-2.87	0.06	1.70	0.98-2.93
AJCC stage
I	-	-	-	-	-	-	-	-	-
II	<0.01	1.33	1.12-1.59	<0.01	3.94	2.61-5.94	<0.01	3.70	2.47-5.55
III	<0.01	1.69	1.47-1.95	<0.01	5.96	4.28-8.29	<0.01	5.77	4.15-8.01
IV	<0.01	4.86	4.27-5.54	<0.01	41.11	30.44-55.50	<0.01	40.73	30.57-54.27

Classification	Cox regression analysis			SD model analysis			CS model analysis
Classification	P-value	HR	95%CI	P-value	HR	95%CI	P-value	HR	95%CI
histology
Papillary	-	-	-	-	-	-	-	-	-
Follicular	<0.01	1.22	1.06-1.40	<0.01	1.44	1.15-1.79	<0.01	1.46	1.20-1.78
Medullary	<0.01	1.51	1.27-1.80	<0.01	1.67	1.33-2.14	<0.01	1.81	1.44-2.27
anaplastic	<0.01	6.74	5.70-7.97	<0.01	6.14	4.79-7.87	<0.01	9.34	7.71-11.31
Surgery
Yes	-	-	-	-	-	-	-	-	-
No	<0.01	1.99	1.71-2.30	<0.01	1.71	1.37-2.12	<0.01	2.06	1.72-2.46
Radiation
Yes	-	-	-	-	-	-	-	-	-
None	<0.01	1.35	1.23-1.48	0.02	1.18	1.02-1.37	<0.01	1.25	1.09-1.44
Chemotherapy
Yes	-	-	-	-	-	-	-	-	-
No/Unknown	<0.01	0.61	0.52-0.74	<0.01	0.53	0.41-0.70	<0.01	0.73	0.61-0.89
Mets at bone
No	-	-	-	-	-	-	-	-	-
Yes	0.14	1.21	0.94-1.57	0.22	1.19	0.90-1.58	0.26	1.17	0.89-1.55
Unknown	0.82	0.90	0.34-2.33	0.87	0.91	0.29-2.84	0.97	0.98	0.33-2.91
Mets at brain
No	-	-	-	-	-	-	-	-	-
Yes	<0.01	1.93	1.24-3.00	0.09	1.72	0.92-3.21	0.01	1.77	1.12-2.80
Unknown	0.41	1.65	0.50-5.46	0.23	2.27	0.60-8.52	0.11	2.92	0.78-10.89
Mets at liver
No	-	-	-	-	-	-	-	-	-
Yes	<0.01	2.23	1.46-3.39	<0.01	2.17	1.41-3.35	<0.01	2.36	1.52-3.66
Unknown	0.78	0.83	0.22-3.11	0.48	0.56	0.12-2.75	0.20	0.38	0.09-1.66
Mets at lung
No	-	-	-	-	-	-	-	-	-
Yes	<0.01	2.31	1.92-2.77	<0.01	2.59	2.07-3.24	<0.01	2.42	1.97-2.98
Unknown	0.99	0.99	0.45-2.17	0.75	1.16	0.46-2.94	0.66	1.21	0.51-2.87

Prognostic factors in patients with thyroid carcinoma: a competing-risks analysis

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusion

Declarations

References

Tables

Status:

Version 1