- Email: [email protected]
Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancer with High-Risk Features
Journal Pre-proof Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancers with High-Risk Features Paritosh Suman, MD FACS, Shantanu N. Razdan, MD, Chi-Hsiung E. Wang, PhD, Mark Tulchinsky, MD, FACNM, Leaque Ahmed, MD, FACS, Richard A. Prinz, MD, FACS, David J. Winchester, MD, FACS PII:
S1072-7515(19)32142-8
DOI:
https://doi.org/10.1016/j.jamcollsurg.2019.09.021
Reference:
ACS 9665
To appear in:
Journal of the American College of Surgeons
Received Date: 1 July 2019 Revised Date:
1 August 2019
Accepted Date: 16 September 2019
Please cite this article as: Suman P, Razdan SN, Wang C-HE, Tulchinsky M, Ahmed L, Prinz RA, Winchester DJ, Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancers with High-Risk Features, Journal of the American College of Surgeons (2019), doi: https://doi.org/10.1016/ j.jamcollsurg.2019.09.021. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc. on behalf of the American College of Surgeons.
1
Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancers with High-Risk Features Paritosh Suman, MD FACSa,b; Shantanu N Razdan, MDc; Chi-Hsiung E Wang, PhDa; Mark Tulchinsky, MD, FACNMd; Leaque Ahmed, MD, FACSb; Richard A Prinz, MD, FACSa; David J Winchester, MD, FACSa a
Department of Surgery, NorthShore University Health System, Evanston, IL Department of Surgery, Wyckoff Heights Medical Center, Brooklyn, NY c Department of Surgery, Harlem Hospital Center, New York, NY d Department of Nuclear Medicine, Penn State University, Milton S Hershey Medical Center, Hershey, PA b
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 105th Annual Clinical Congress, Scientific Forum, San Francisco, CA, October 2019. Corresponding Author Address: Paritosh Suman, MD, FACS Department of Surgery NorthShore University Health System 2650 Ridge Avenue, Walgreen Suite 2507 Evanston, IL 60201 Tel: 847-570-1316 Fax: 847-570-2930 Email: [email protected] Running Head: Papillary Thyroid Cancer and Thyroid Lobectomy
2
Abstract Background: Thyroid lobectomy (TL) has been proposed as definitive surgical treatment for papillary thyroid cancers (PTC) up to 4 cm. This study evaluates the use and appropriateness of TL for T1b and T2 PTC. Study Design: The National Cancer Database was interrogated for adult patients having TL for T1b-T2 PTC between 2004-2014. Patients who should have undergone total thyroidectomy (TT) instead of lobectomy based on high-risk tumor features were identified. The two groups were compared for clinical and demographic characteristics, and overall survival. Results: Of 8083 patients undergoing lobectomy, 1552 patients had high-risk features and should have undergone TT. These included 194 with cN1, 571 with pN1, 307 with LVI, 645 with ETE, 567 with positive margins, 42 with poorly-differentiated PTC and 25 with M1 disease. At 10 years of follow up, 92.4% of appropriate lobectomy (aTL) patients were alive compared to 88.5% of inappropriate lobectomy (iTL) patients (p<.001). On univariate and multivariable Cox survival analysis, age >45 years, male gender, comorbidities, government or no insurance, low income and tumor size >2 cm were associated with poorer survival (all p<.05). Thyroid lobectomy patients with high-risk features had significantly higher mortality on unadjusted (HR 1.98, 95% CI 1.52-2.59, p<.001) and adjusted survival analysis (HR 1.97, 95% CI 1.51-2.58, p<.001). Total thyroidectomy with radioiodine treatment had improved OS in comparison to iTL (HR .65, 95% CI .51-.83, p<.001). Conclusion: A substantial number of patients (19.2%) with tumor size >1 cm and high-risk features undergo thyroid lobectomy for PTC. Exclusion of high-risk features is important when adopting lobectomy as the definitive surgical therapy for T1b and T2 PTC since they have a potential adverse effect on long-term survival.
3
Keywords: Thyroid Cancer; Papillary Thyroid Cancer; Thyroid Lobectomy; National Cancer Database; Overall Survival
Abbreviations ATA, American Thyroid Association aTL, Appropriate Thyroid Lobectomy CT, Completion Thyroidectomy ETE, Extra Thyroidal Extension HR, Hazard Ratio iTL, Inappropriate Thyroid Lobectomy KM, Kaplan-Meier LVI, Lymphovascular Invasion NCDB, National Cancer Database OR, Odds Ratio OS, Overall Survival PDTC, Poorly Differentiated Thyroid Cancer PTC, Papillary Thyroid Cancer RAI, Radioactive Iodine TL, Thyroid Lobectomy TT, Total Thyroidectomy TT+RAIT, Total Thyroidectomy with Radioactive Iodine Treatment
4
Introduction The incidence of thyroid cancer has been increasing in the US with papillary thyroid cancer (PTC) accounting for most of the new diagnoses (1). While survival from PTC is excellent, there are certain tumor and clinicodemographic features which adversely affect long-term survival. Excellent overall survival of patients with PTC contributes to the challenge of determining optimal therapy, including the pivotal question of initial surgical extent. The surgical procedure choice is usually made in the course of preoperative consultation between the surgeon and the well-informed patient. Total or near total thyroidectomy (TT) has been considered the standard procedure for PTC greater than 1 cm in size (2). Recently, 2015 American Thyroid Association (ATA) guidelines reconsidered thyroid lobectomy (TL) as an acceptable option for some of these cancers (2). This publication states that TL is specifically indicated for tumors 1 – 4 cm (T1b – T2) in size and without any of the high-risk features. The proposed advantages of TL include fewer complications, shorter operative times and better health related quality of life (3). If postTL pathology reveals high-risk feature(s), a completion thyroidectomy (CT) remains an available option. A recent study using the NSQIP database reported that since the publication of 2015 ATA guidelines the rate of TL for PTC has been increasing in the US (4). With increasing use of TL (5), it is important with a T1b – T2 PTC to rule out high-risk features preoperatively and to recommend a completion thyroidectomy if they are identified postoperatively. Thus, appropriate preoperative evaluation and monitoring of pathology results for any high-risk features are vital for improving oncological outcomes with these PTC. As TL is being adopted for PTC with low-risk features, there is concern that patients who have high-risk features with PTC are undergoing a TL as their definitive treatment. Inappropriate
5
TL (iTL) is possible with either inadequate preoperative evaluation or failure at post-TL recognition of high-risk features on surgical pathology. We hypothesized that iTL may lead to a decrease in overall survival (OS). Information on prevalence of iTL in the US practice is lacking, but measuring and minimizing it would be an important metric of the surgical quality. The primary aim of our study was to identify the prevalence of iTL. The secondary aim was to assess the effect of iTL on the overall survival of patients with T1b-T2 PTC.
Methods Patient Population The American College of Surgeons has executed a Business Associate Agreement that includes a data use agreement with each of its Commission on Cancer accredited hospitals. The NCDB is a nationwide, facility-based, comprehensive clinical surveillance resource oncology data set that currently captures 70% of all newly diagnosed malignancies in the US annually. The NCDB Thyroid participant user file was used to obtain data on PTC diagnosed between 2004 and 2014. Inclusion criteria consisted of patients with PTC and tumor size ranging from 1.1 to 4 cm (clinical T1b – T2) who underwent thyroid lobectomy as their definitive operation. Using the International Classification of Diseases for Oncology, 3rd edition, the following histology codes were included: 8050, 8260, 8262, 8340-4 and 8450. Exclusion criteria consisted of tumors < 1 cm and > 4 cm and those patients who had TT as their definitive operation. Baseline patient variables such as age, gender, race, comorbidities (Charlson Deyo Comorbidity Index), insurance and income status were recorded. Facility level data recorded included type of hospital which was reclassified into academic and non-academic cancer centers. Clinical data noted was tumor size (T), clinical nodal status (cN), pathological nodal status (pN), metastasis (M),
6
lymphovascular invasion (LVI), extra thyroidal extension (ETE), tumor margin status and degree of tumor differentiation. Survival data was recorded as overall survival at 5 and 10 years postoperatively. Data about preoperative imaging, tumor recurrence, disease specific survival, number and dosage of radioactive iodine (RAI) treatments, and serum thyroglobulin levels were not available in the NCDB. Patients who had high-risk features were placed in the study group. High-risk features included presence of clinical or pathological node positive tumors, metastasis, ETE, LVI, positive margins and poorly differentiated tumors. These patients were grouped as iTL. The remaining patients without high-risk features were grouped as appropriate TL (aTL).
Statistics The aTL and iTL groups were compared based on patient, facility, clinical and survival data. Descriptive statistical analysis for continuous and categorical variables was done using t-test and chi-square test respectively. A p value of less than .05 was considered as statistically significant. A Kaplan-Meier (KM) analysis was performed and survival distributions were compared between the aTL and iTL groups using the log-rank test. Univariate Cox regression analysis was done for each baseline variable. Factors noted to be statistically significant were added to a multivariable Cox proportional hazards model with inappropriate lobectomy as the independent variable. Subsequently, a group of patients who had total thyroidectomy with RAI (TT+RAIT) for T1b-2 PTC with high-risk features was identified and was compared with patients in iTL group.
Propensity score matching
7
In order to reduce the bias of confounding variables between the two groups a propensity score matching was performed to identify similar patients for comparison. A propensity score model was generated to assess the factors that could influence the performance of iTL. The propensity score was calculated by fitting a logistic regression model with type of thyroidectomy as the outcome variable and clinically relevant variables as the predictors. Age, gender, race, comorbidities, hospital type, income, insurance status and tumor size were considered for matching. A propensity score matching was subsequently performed where all patients in iTL group were compared with a same matched number of aTL patients. Similar matching was performed for patients in iTL and TT+RAIT groups. Survival analysis was subsequently performed in the propensity-score matched datasets.
Results A total of 8083 patients were identified in this study. Of 8083 patients, 6531 had an aTL (81%) and 1552 patients had an iTL (19%). High risk features in the iTL group included cN1 (n=194), ETE (n=645), positive margins (n=567), M1 disease (n=25), pN1 (n=571), LVI (n=307), and poorly differentiated cancer (n=42). 12% (n=1001) and 5% (n=368) of the patients had more than one or more than two high-risk features respectively. The mean age (standard deviation) of patients in the aTL and iTL group was 47.4 (14.8) and 47.8 (15.9) years respectively (Table 1). The proportion of patients less than and more than 45 years of age was similar between the two groups (p=.691). The number of female compared to male patients was higher in each group (p=.0007). Compared to those in the aTL group, patients who had iTL had more comorbidities (16% vs 13%, p=.027), were more likely to be of White-Hispanic race (8% vs 6%, p<.0001), and have government insurance (27% vs 23%, p=.001). There was no difference in income status
8
(p=.777) and facility type (p=.064) between the two groups. Patients who had an iTL were more likely to have a tumor size between 2.1 to 4 cm (p=.026). On adjusted logistic regression analysis, the following factors were associated with iTL: presence of comorbidities (OR 1.26, 95% CI 1.96 – 1.50, p=.011), government insurance (OR 1.34, 95% CI 1.16 – 1.57, p<.007), no insurance (OR 1.53, 95% CI 1.11 – 2.10, p=.009), tumor size 2.1 to 4 cm (OR 1.26, 95% CI 1.10 – 1.45, p=.001) and White-Hispanic race (OR 1.58, 95% CI 1.20 – 2.09, p=.001).
Survival Analysis Mean overall survival in the iTL group was lower compared to that in the aTL group (9.81 years vs 10.75 years). Patients in the iTL group had significantly lower 5-year and 10-year overall survival as compared to those in the aTL group (94.7% vs 97.5% and 88.5% vs 92.4% respectively, log rank p<.001). On propensity score matching analysis, patients in the iTL group still had significantly lower 5-year and 10-year overall survival as compared to those in the aTL group (94.7% vs 96.5% and 88.5% vs 90.5% respectively, log-rank p=.004). (Table 2, Figure 1) On univariate Cox analysis, iTL was associated with poor survival as compared to aTL (HR 1.98, 95% CI 1.52 – 2.59, p<0.001). Other factors associated with adverse survival were age > 45 years (p<.001), male gender (p<.001), presence of comorbidities (p<.001), government insurance (p<0.001) and low-income status (p<.001). iTL group consisted of patients with only high-risk features, therefore these high-risk features were not included again as independent variables for adjusted survival analysis. On multivariable Cox survival analysis, overall survival was worse in patients with iTL (HR 1.97, 95% CI 1.51 – 2.58, p<.001), age > 45 years (HR 4.63, 95% CI 3.06 – 7.00, p<.001), male gender (HR 1.53, 95% CI 1.18 – 2.00, p<.001), presence of
9
comorbidities (HR 1.72, 95% CI 1.31 – 2.26, p<.001), government insurance (HR 4.88, 95% CI 3.71 – 6.43, p <.001) and low income status (HR 1.60, 95% CI 1.19 – 2.14, p=.002). (Table 3) In a separate subgroup analysis of patients without metastatic disease but other high-risk factors, iTL was still associated with a reduced 5-year (95.3% vs 97.5%) and 10-year (89.1% vs 92.4%) overall survival compared to patients undergoing aTL (log rank p<.001). A subsequent comparison between patients in iTL and TT+RAIT groups was performed. (Table 4) In comparison to iTL group (n=1552), patients in TT+RAIT group (n=36938) who had total thyroidectomy and RAI for similar tumors with high-risk features had significantly improved 5-year (97.1% vs 94.7%) and 10-year (92.3% vs 88.5%) overall survival (log rank p<.001). (Figure 2) Total thyroidectomy with subsequent RAI treatment was associated with improved survival on Cox univariate (HR .56, 95% CI .44 – .70, p<.001) and adjusted multivariable survival analysis (HR .65, 95% CI .51 – .83, p<.001). On propensity score matching analysis, patients in the TT+RAIT group maintained survival advantage as compared to those in the iTL group (5 years- 96.5% vs 94.7% and 10 years- 90.9% vs 88.5% respectively, log-rank p=.031). (Table 5)
Discussion We conducted a statistically well-powered retrospective study analyzing patients with T1b and T2 PTC in the NCDB treated from 2004 to 2014. These patients were treated before publication of the pivotal 2015 ATA guidelines where lobectomy was for the first time formally recommended as an acceptable alternative to total thyroidectomy for PTC with a tumor size 1.1 to 4 cm. However, there were 1552 patients had a lobectomy for these tumors when total thyroidectomy should have been the recommended operation. Applying the ATA guidelines, we
10
identified that approximately 19% of the patients had a iTL as the definitive operation for their T1b – T2 PTC with high-risk features. In theory these patients should have had an initial TT or a subsequent completion thyroidectomy. This may be due to inadequate preoperative evaluation leading to inappropriate patient selection, lack of review of the postoperative pathologic results identifying high-risk features or patients who refused to have a completion thyroidectomy or who were lost to follow up. The high incidence of thyroid lobectomy in patients with high-risk features suggests there is a need to improve and standardize the preoperative and postoperative evaluation of these patients and to develop protocols to ensure the appropriate operation is performed. A detailed imaging of thyroid nodule and cervical lymph nodes, availability of dedicated endocrine cytopathologist, assessment of vocal cord mobility, discussion in multidisciplinary tumor board, and a system to track and follow patients could be considered for such management protocols. Individual high-risk features have been shown to be associated with increased recurrence and reduced survival in PTC. In a study of 47,902 patients with PTC, Adam et al. concluded that presence of clinical lymph node metastasis was linked with poorer overall survival (HR 1.32 95% CI 1.04-1.67, p=.021) (6). Those authors advised preoperative screening for lymph node metastases in all patients with PTC. Many patients with PTC can have very long periods of progression-free survival even in the presence of distant metastases. This was shown by a single institutional study of 3664 patients from Memorial Sloan Kettering Cancer Center where a 77.6% five-year survival could be achieved (7). However, such excellent survival outcomes require complete extirpation of thyroid tissue with subsequent radioiodine therapy. In a study from the NCDB, Youngswirth et al. concluded that both minimal and extensive ETE were associated with reduced overall survival in patients with differentiated thyroid cancers (8). This
11
finding carries significance as extensive ETE could potentially be identified preoperatively as reported by Kuo et al. They showed that a detailed sonographic assessment of ETE led to a reduced rate of potentially avoidable total thyroidectomy from 57% to 31% (9). With this approach they were able to reliably exclude ETE preoperatively in most patients. Pathology specimens must be reviewed to identify any high-risk histological features when present. Pathologically positive neck lymph nodes are a major determinant in PTC staging and affect the need for subsequent RAI treatment and cancer surveillance of PTC. Therefore, detection of positive lymph nodes in the thyroid pathology specimens requires that a completion thyroidectomy be considered to enable RAI treatment and to decrease the recurrence rate (2). Although rare, poorly differentiated thyroid cancers (PTDC) are a major survival determinant of in nonanaplastic follicular cell derived thyroid cancers. Total thyroidectomy is the recommended surgical treatment for these patients (10). Ideally, PDTC should be identified on preoperative evaluation. When PDTC is found after thyroid lobectomy, a completion thyroidectomy is strongly advised. Similarly, microscopic and macroscopic positive margins have been found to be associated with reduced overall survival in patients with PTC (11). This is a high-risk feature for recurrence of PTC so a completion thyroidectomy should be considered. Additionally, LVI also affects survival independent of other high-risk features as described by Pontius et al. (12). Presence of LVI in the final pathology specimen ideally warrants a completion thyroidectomy to enable RAI treatment. RAI treatment is associated with improved survival in these patients (12). Patients with postoperatively detected adverse tumor features must be tracked and advised of the seriousness of these features to ensure an optimal oncologic outcome. While the ATA guidelines recommend either a total/near total thyroidectomy or a lobectomy for PTC 1 – 4 cm without extra thyroidal extension and without clinical evidence of
12
lymph node metastasis, much of the data supporting use of lobectomy for PTC has been derived from selective cancer centers. The 2015 ATA guidelines were primarily based on supporting evidence for unilateral thyroid surgery for T1-2 PTC from two large volume academic cancer centers from US and Japan (13, 14). These centers tend to have rigorous preoperative protocols that lead to better patient selection as well as postoperative surveillance protocols to identify patients that meet the criteria for completion thyroidectomy. Smaller community programs often lack a multi-disciplinary perioperative oncology team so the use of lobectomies in these centers may not achieve the favorable results of academic centers. Patients that underwent iTL had significantly poorer overall survival at 5 and 10 years after their operation than those who underwent aTL. This is an expected finding as patients in the iTL group had high-risk features that may not have been accounted for in the operative decisionmaking process and resulted in the poorer prognosis. Propensity score matching was done to ensure that similar patients were compared between both groups. The negative effect of iTL on overall 5 and 10-year survival was still observed. Even after adjusting for baseline confounding variables, patients who had aTL were twice as likely to survive as those who underwent iTL. Even though the number of patients with metastatic disease was low in our study due to a concern that it might disproportionately affect overall survival, we decided to exclude these patients and do a separate subgroup analysis. The result confirmed the main finding that overall survival was better for patients in aTL group as compared to the iTL group. Overall survival showed significant improvement when patients with > 1 to 4 cm PTC and high-risk features underwent total thyroidectomy with subsequent RAI. This finding was not in agreement with results of the 2014 study by Adam et al. (15) They found no survival difference between patients undergoing thyroid lobectomy and total thyroidectomy for T1b-T2 PTC. Selection of patients
13
with only high-risk features, and comparison between patients undergoing either inappropriate lobectomy or appropriate total thyroidectomy with RAI were the major differences in our study compared to that by Adam et al. Our results emphasize the need for a thorough preoperative assessment of risk factors for patients diagnosed with PTC. Once high-risk features are identified pre- or post-operatively, these patients should be advised that TT or completion thyroidectomy, respectively may provide a better outcome. To our knowledge this is the first study that has reported the high proportion of patients undergoing inappropriate lobectomies for PTC in the US and its negative effect on overall survival. We also used propensity score matching to eliminate the effect of confounders on our results. The findings of our study are important for thyroid surgeons to improve their patient outcomes. Results of our study can also be used to better inform patients with PTC when making their decision regarding the type of thyroidectomy. Our study does have several limitations. The NCDB does not have data on disease specific survival. However, PTC is a slow growing cancer with excellent long-term survival and we believe that overall survival is a reliable proxy for disease specific survival. There is also no information on patient preferences or intra- or post-operative decision making that could have altered the choice between TL and TT in these patients. Additionally, the NCDB lacks data about recurrence of PTC, hence, the cause of better survival remains presumptive.
Conclusions The results of our study show that a substantial fraction of patients with T1b-2 PTC and high-risk features underwent iTL instead of TT+RAIT. Our results also suggest that iTL has a significant negative effect on long-term survival of these patients. Surgeons performing TL and TT now
14
have an opportunity to evaluate their individual statistics. The focus should be on improving effectiveness of pre- and post-operative processes with a goal of minimizing the rate of iTL in an individual’s surgical practice. Thyroid surgeons should adhere to and carefully individualize the applicability of treatment guidelines when determining the type of thyroidectomy for their patients with PTC.
15
References 1. James BC, Timsina L, Graham R, et al. Changes in total thyroidectomy versus thyroid lobectomy for papillary thyroid cancer during the past 15 years. Surgery 2019;166(1):4147. 2. Haugen BR, Alexander EK, Bible KC, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: The American thyroid association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016;26(1):1-133. 3. Nickel B, Tan T, Cvejic E, et al. Health-related quality of life after diagnosis and treatment of differentiated thyroid cancer and association with type of surgical treatment. JAMA Otolaryngol Head Neck Surg. 2019 Jan 17. [Epub ahead of print] 4. Ullmann TM, Gray KD, Stefanova D, et al. The 2015 American thyroid association guidelines are associated with an increasing rate of hemithyroidectomy for thyroid cancer. Surgery 2019 May 2. [Epub ahead of print] 5. Vargas-Pinto S, Romero Arenas MA. Lobectomy compared to total thyroidectomy for low-risk papillary thyroid cancer: A systematic review. J Surg Res 2019;242:244-251. 6. Adam MA, Pura J, Goffredo P, et al. Presence and number of lymph node metastases are associated with compromised survival for patients younger than age 45 years with papillary thyroid cancer. J Clin Oncol 2015;33(21):2370-2375. 7. Wang LY, Palmer FL, Nixon IJ, et al. Multi-organ distant metastases confer worse disease-specific survival in differentiated thyroid cancer. Thyroid 2014;24(11):15941599.
16
8. Youngwirth LM, Adam MA, Scheri RP, et al. Extrathyroidal extension is associated with compromised survival in patients with thyroid cancer. Thyroid 2017;27(5):626-631. 9. Kuo EJ, Thi WJ, Zheng F, et al. Individualizing surgery in papillary thyroid carcinoma based on a detailed sonographic assessment of extrathyroidal extension. Thyroid 2017;27(12):1544-1549. 10. Ibrahimpasic T, Ghossein R, Carlson DL, et al. Outcomes in patients with poorly differentiated thyroid carcinoma. J Clin Endocrinol Metab 2014;99(4):1245-1252. 11. Youngwirth LM, Adam MA, Scheri RP, et al. Patients treated at low-volume centers have higher rates of incomplete resection and compromised outcomes: Analysis of 31,129 patients with papillary thyroid cancer. Ann Surg Oncol 2016;23(2):403-409. 12. Pontius LN, Youngwirth LM, Thomas SM, et al. Lymphovascular invasion is associated with survival for papillary thyroid cancer. Endocr Relat Cancer 2016;23(7):555-562. 13. Nixon IJ, Ganly I, Patel SG, et al. Thyroid lobectomy for treatment of well differentiated intrathyroid malignancy. Surgery 2012;151(4):571-579. 14. Matsuzu K, Sugino K, Masudo K, et al. Thyroid lobectomy for papillary thyroid cancer: long-term follow-up study of 1,088 cases. World J Surg 2014;38(1):68-79. 15. Adam MA, Pura J, Gu L, et al. Extent of surgery for papillary thyroid cancer is not associated with survival: an analysis of 61,775 patients. Ann Surg 2014;260(4):601-605.
17
Table 1. Comparison of Demographic and Clinical Characteristics of the Patients Having an Appropriate or an Inappropriate Thyroid Lobectomy Appropriate Inappropriate Variable All thyroid thyroid p Value lobectomy lobectomy Age Mean years ± 47.5 ± 15.1 47.4 ±14.8 47.8 ± 15.9 .108 SD ≤ 45 y, n (%) 3,612 (45) 2,911 (45) 701 (45) .691 > 45 y, n (%) 4,471 (55) 3,620 (55) 851 (55) Sex, n (%) Male 1,743 (22) 1,369 (21) 374 (24) .007* Female 6,340 (78) 5,162 (79) 1,178 (76) Comorbidity, n (%) No 6,971 (86) 5,660 (87) 1,311 (84) .027* Yes 1,112 (14) 871 (13) 241 (16) Race, n (%) White non5,884 (72) 4,756 (73) 1,088 (70) <.001* Hispanic White 495 (6) 369 (6) 126 (8) Hispanic African 656 (8) 559 (9) 97 (6) American Other 1,088 (14) 847 (12) 241 (16) Insurance, n (%) None 383 (5) 298 (5) 85 (5) .001* Private 5,775 (71) 4,725 (72) 1,050 (68) Government 1,925 (24) 1,508 (23) 417 (27) Income, n (%) Low 1,150 (14) 933 (14) 217 (14) .777 High 6,884 (85) 5,556 (86) 1,328 (86) Cancer center, n (%) Academic 2,037 (37) 1,618 (37) 419 (41) .020* Non3,399 (63) 2,788 (63) 611 (59) academic Tumor size, cm Mean size ± 2.13 ± 0.81 2.14 ± 0.8 2.18 ± 0.84 .026* SD 1.1 - 2, n (%) 4,554 (56) 3,719 (57) 835 (54) 2.1 - 4, n (%) 3,529 (44) 2,812 (43) 717 (46) *Significant.
18
Table 2. Survival Analysis for Papillary Thyroid Cancer Patients Having an Inappropriate Thyroid Lobectomy Compared to Those Having an Appropriate Thyroid Lobectomy Overall survival Patient group Mean ± SE, y
5-year, %
10-year, %
10.75 ± .03
97.5
92.4
p Value
All patients Appropriate lobectomy
<.001* Inappropriate lobectomy
9.81 ± .07
94.7
88.5
10.39 ± .06
96.5
90.5
Propensity score matched patients Appropriate lobectomy
.004* Inappropriate lobectomy *Significant.
9.81 ± .07
94.7
88.5
19
Table 3. Univariate and Multivariable Cox Regression Analysis for Factors Associated with Overall Survival Univariate Multivariable Predictor HR 95% CI p Value HR 95% CI p Value Lobectomy Appropriate 1 1 Inappropriate 1.98 1.52 - 2.59 <.001* 1.97 1.51 - 2.58 <.001* Age ≤ 45 y 1 1 > 45 y 7.56 5.04 - 11.34 <.001* 4.63 3.06 - 7.00 <.001* Sex Female 1 1 Male 1.79 1.38 - 2.32 <.001* 1.53 1.18 - 2.00 .001* Comorbidity No 1 1 Yes 3.02 2.32 - 3.19 <.001* 1.72 1.31 - 2.26 .001* Race White non-Hispanic 1 1 Other 0.37 0.22 - 0.63 <.001* 0.37 0.21 - 0.63 <.001* White Hispanic 0.47 0.23 - 0.95 .037* 0.44 0.22 - 0.90 .025* African American 1.27 0.85 - 1.89 .247 0.85 0.56 - 1.30 .446 Insurance Private 1 1 None 2.47 1.35 - 4.53 .004* 2.51 1.36 - 4.63 .003* Government 7.37 5.64 - 9.60 <.001* 4.88 3.71 - 6.43 <.001* Income High 1 1 Low 2.01 1.52 - 2.66 <.001* 1.60 1.19 - 2.14 .002* Cancer center Academic 1 Non-academic 0.99 0.76 - 1.29 .920 Tumor size, cm 1.1 - 2 1 2.1 - 4 1.17 0.91 - 1.49 .216 *Significant. HR, hazard ratio.
20
Table 4. Comparison of Demographic and Clinical Characteristics of the Patients Having an Inappropriate Thyroid Lobectomy or Total Thyroidectomy with Radioactive Iodine Treatment Variable All iTL TT+RAIT p Value Age Mean size ± SD 45.9 ± 14.9 48.1 ±16.1 45.8 ± 14.9 <.001* ≤ 45 y, n (%) 18,670 (48) 701 (45) 17,969 (49) .004* > 45 y, n (%) 19,820 (52) 851 (55) 18,969 (51) Sex, n (%) Male 9,977 (26) 374 (24) 9,603 (26) .050* Female 28,513 (72) 1,178 (76) 27,335 (74) Comorbidity, n (%) No 33,096 (86) 1,311 (84) 31,785 (86) .043* Yes 5,394 (14) 241 (16) 5,153 (14) Race, n (%) White non-Hispanic 27,084 (70) 1,088 (70) 25,996 (65) <.001* White Hispanic 4,023 (10) 126 (8) 3,897 (11) African American 1,651 (4) 97 (6) 1,554 (4) Other 5,732 (16) 241 (16) 5,491 (20) Insurance, n (%) None 1,644 (4) 85 (5) 1559 (4) <.001* Private 29,028 (75) 1,050 (72) 27,978 (76) Government 7,818 (21) 417 (23) 7,401 (20) Income, n (%) Low 4,380 (11) 217 (14) 4,163 (11) .001* High 33,845 (89) 1,328 (86) 32,517 (89) Cancer center, n (%) Academic 10,974 (45) 419 (41) 10,555 (45) .003* Non-academic 13,488 (55) 611 (59) 12,877 (55) Tumor size, cm Mean size ± SD 2.15 ± 0.8 2.18 ± 0.84 2.15 ± 0.8 .329 1.1 - 2, n (%) 20,932 (54) 835 (54) 20,097 (54) 2.1 - 4, n (%) 17,558 (46) 717 (46) 16,841 (46) *Significant. iTL, inappropriate thyroid lobectomy; TT+RAIT, total thyroidectomy with radioactive iodine treatment.
21
Table 5. Survival Analysis for Papillary Thyroid Cancer Patients Having an Inappropriate Thyroid Lobectomy Compared to Those Having a Total Thyroidectomy with Radioactive Iodine Treatment Overall survival Patient group
5-year, %
10-year, %
p Value
Hazard ratio
95% CI
p Value
Inappropriate lobectomy
94.7
88.5
<.001*
1
Multivariable
<.001*
Total thyroidectomy
97.1
92.3
.65
.51 - .83
94.7
88.5
96.5
90.9
All patients
Propensity score matched patients Inappropriate lobectomy Total thyroidectomy *Significant.
.031*
22
Figure Legends Figure 1. Kaplan-Meier plot comparing survival between appropriate lobectomy (aTL) and inappropriate lobectomy (iTL) groups. Figure 2. Kaplan-Meier plot comparing survival between inappropriate lobectomy (iTL) and total thyroidectomy with radioactive iodine treatment (TT+RAIT) groups.
23
Precis A total of 19.2% patients with T1b-T2 (1.1 – 4 cm in size) papillary thyroid cancer had high-risk features and underwent inappropriate thyroid lobectomy with an adverse impact on overall survival.
S1072-7515(19)32142-8
DOI:
https://doi.org/10.1016/j.jamcollsurg.2019.09.021
Reference:
ACS 9665
To appear in:
Journal of the American College of Surgeons
Received Date: 1 July 2019 Revised Date:
1 August 2019
Accepted Date: 16 September 2019
Please cite this article as: Suman P, Razdan SN, Wang C-HE, Tulchinsky M, Ahmed L, Prinz RA, Winchester DJ, Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancers with High-Risk Features, Journal of the American College of Surgeons (2019), doi: https://doi.org/10.1016/ j.jamcollsurg.2019.09.021. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc. on behalf of the American College of Surgeons.
1
Thyroid Lobectomy for T1b-T2 Papillary Thyroid Cancers with High-Risk Features Paritosh Suman, MD FACSa,b; Shantanu N Razdan, MDc; Chi-Hsiung E Wang, PhDa; Mark Tulchinsky, MD, FACNMd; Leaque Ahmed, MD, FACSb; Richard A Prinz, MD, FACSa; David J Winchester, MD, FACSa a
Department of Surgery, NorthShore University Health System, Evanston, IL Department of Surgery, Wyckoff Heights Medical Center, Brooklyn, NY c Department of Surgery, Harlem Hospital Center, New York, NY d Department of Nuclear Medicine, Penn State University, Milton S Hershey Medical Center, Hershey, PA b
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 105th Annual Clinical Congress, Scientific Forum, San Francisco, CA, October 2019. Corresponding Author Address: Paritosh Suman, MD, FACS Department of Surgery NorthShore University Health System 2650 Ridge Avenue, Walgreen Suite 2507 Evanston, IL 60201 Tel: 847-570-1316 Fax: 847-570-2930 Email: [email protected] Running Head: Papillary Thyroid Cancer and Thyroid Lobectomy
2
Abstract Background: Thyroid lobectomy (TL) has been proposed as definitive surgical treatment for papillary thyroid cancers (PTC) up to 4 cm. This study evaluates the use and appropriateness of TL for T1b and T2 PTC. Study Design: The National Cancer Database was interrogated for adult patients having TL for T1b-T2 PTC between 2004-2014. Patients who should have undergone total thyroidectomy (TT) instead of lobectomy based on high-risk tumor features were identified. The two groups were compared for clinical and demographic characteristics, and overall survival. Results: Of 8083 patients undergoing lobectomy, 1552 patients had high-risk features and should have undergone TT. These included 194 with cN1, 571 with pN1, 307 with LVI, 645 with ETE, 567 with positive margins, 42 with poorly-differentiated PTC and 25 with M1 disease. At 10 years of follow up, 92.4% of appropriate lobectomy (aTL) patients were alive compared to 88.5% of inappropriate lobectomy (iTL) patients (p<.001). On univariate and multivariable Cox survival analysis, age >45 years, male gender, comorbidities, government or no insurance, low income and tumor size >2 cm were associated with poorer survival (all p<.05). Thyroid lobectomy patients with high-risk features had significantly higher mortality on unadjusted (HR 1.98, 95% CI 1.52-2.59, p<.001) and adjusted survival analysis (HR 1.97, 95% CI 1.51-2.58, p<.001). Total thyroidectomy with radioiodine treatment had improved OS in comparison to iTL (HR .65, 95% CI .51-.83, p<.001). Conclusion: A substantial number of patients (19.2%) with tumor size >1 cm and high-risk features undergo thyroid lobectomy for PTC. Exclusion of high-risk features is important when adopting lobectomy as the definitive surgical therapy for T1b and T2 PTC since they have a potential adverse effect on long-term survival.
3
Keywords: Thyroid Cancer; Papillary Thyroid Cancer; Thyroid Lobectomy; National Cancer Database; Overall Survival
Abbreviations ATA, American Thyroid Association aTL, Appropriate Thyroid Lobectomy CT, Completion Thyroidectomy ETE, Extra Thyroidal Extension HR, Hazard Ratio iTL, Inappropriate Thyroid Lobectomy KM, Kaplan-Meier LVI, Lymphovascular Invasion NCDB, National Cancer Database OR, Odds Ratio OS, Overall Survival PDTC, Poorly Differentiated Thyroid Cancer PTC, Papillary Thyroid Cancer RAI, Radioactive Iodine TL, Thyroid Lobectomy TT, Total Thyroidectomy TT+RAIT, Total Thyroidectomy with Radioactive Iodine Treatment
4
Introduction The incidence of thyroid cancer has been increasing in the US with papillary thyroid cancer (PTC) accounting for most of the new diagnoses (1). While survival from PTC is excellent, there are certain tumor and clinicodemographic features which adversely affect long-term survival. Excellent overall survival of patients with PTC contributes to the challenge of determining optimal therapy, including the pivotal question of initial surgical extent. The surgical procedure choice is usually made in the course of preoperative consultation between the surgeon and the well-informed patient. Total or near total thyroidectomy (TT) has been considered the standard procedure for PTC greater than 1 cm in size (2). Recently, 2015 American Thyroid Association (ATA) guidelines reconsidered thyroid lobectomy (TL) as an acceptable option for some of these cancers (2). This publication states that TL is specifically indicated for tumors 1 – 4 cm (T1b – T2) in size and without any of the high-risk features. The proposed advantages of TL include fewer complications, shorter operative times and better health related quality of life (3). If postTL pathology reveals high-risk feature(s), a completion thyroidectomy (CT) remains an available option. A recent study using the NSQIP database reported that since the publication of 2015 ATA guidelines the rate of TL for PTC has been increasing in the US (4). With increasing use of TL (5), it is important with a T1b – T2 PTC to rule out high-risk features preoperatively and to recommend a completion thyroidectomy if they are identified postoperatively. Thus, appropriate preoperative evaluation and monitoring of pathology results for any high-risk features are vital for improving oncological outcomes with these PTC. As TL is being adopted for PTC with low-risk features, there is concern that patients who have high-risk features with PTC are undergoing a TL as their definitive treatment. Inappropriate
5
TL (iTL) is possible with either inadequate preoperative evaluation or failure at post-TL recognition of high-risk features on surgical pathology. We hypothesized that iTL may lead to a decrease in overall survival (OS). Information on prevalence of iTL in the US practice is lacking, but measuring and minimizing it would be an important metric of the surgical quality. The primary aim of our study was to identify the prevalence of iTL. The secondary aim was to assess the effect of iTL on the overall survival of patients with T1b-T2 PTC.
Methods Patient Population The American College of Surgeons has executed a Business Associate Agreement that includes a data use agreement with each of its Commission on Cancer accredited hospitals. The NCDB is a nationwide, facility-based, comprehensive clinical surveillance resource oncology data set that currently captures 70% of all newly diagnosed malignancies in the US annually. The NCDB Thyroid participant user file was used to obtain data on PTC diagnosed between 2004 and 2014. Inclusion criteria consisted of patients with PTC and tumor size ranging from 1.1 to 4 cm (clinical T1b – T2) who underwent thyroid lobectomy as their definitive operation. Using the International Classification of Diseases for Oncology, 3rd edition, the following histology codes were included: 8050, 8260, 8262, 8340-4 and 8450. Exclusion criteria consisted of tumors < 1 cm and > 4 cm and those patients who had TT as their definitive operation. Baseline patient variables such as age, gender, race, comorbidities (Charlson Deyo Comorbidity Index), insurance and income status were recorded. Facility level data recorded included type of hospital which was reclassified into academic and non-academic cancer centers. Clinical data noted was tumor size (T), clinical nodal status (cN), pathological nodal status (pN), metastasis (M),
6
lymphovascular invasion (LVI), extra thyroidal extension (ETE), tumor margin status and degree of tumor differentiation. Survival data was recorded as overall survival at 5 and 10 years postoperatively. Data about preoperative imaging, tumor recurrence, disease specific survival, number and dosage of radioactive iodine (RAI) treatments, and serum thyroglobulin levels were not available in the NCDB. Patients who had high-risk features were placed in the study group. High-risk features included presence of clinical or pathological node positive tumors, metastasis, ETE, LVI, positive margins and poorly differentiated tumors. These patients were grouped as iTL. The remaining patients without high-risk features were grouped as appropriate TL (aTL).
Statistics The aTL and iTL groups were compared based on patient, facility, clinical and survival data. Descriptive statistical analysis for continuous and categorical variables was done using t-test and chi-square test respectively. A p value of less than .05 was considered as statistically significant. A Kaplan-Meier (KM) analysis was performed and survival distributions were compared between the aTL and iTL groups using the log-rank test. Univariate Cox regression analysis was done for each baseline variable. Factors noted to be statistically significant were added to a multivariable Cox proportional hazards model with inappropriate lobectomy as the independent variable. Subsequently, a group of patients who had total thyroidectomy with RAI (TT+RAIT) for T1b-2 PTC with high-risk features was identified and was compared with patients in iTL group.
Propensity score matching
7
In order to reduce the bias of confounding variables between the two groups a propensity score matching was performed to identify similar patients for comparison. A propensity score model was generated to assess the factors that could influence the performance of iTL. The propensity score was calculated by fitting a logistic regression model with type of thyroidectomy as the outcome variable and clinically relevant variables as the predictors. Age, gender, race, comorbidities, hospital type, income, insurance status and tumor size were considered for matching. A propensity score matching was subsequently performed where all patients in iTL group were compared with a same matched number of aTL patients. Similar matching was performed for patients in iTL and TT+RAIT groups. Survival analysis was subsequently performed in the propensity-score matched datasets.
Results A total of 8083 patients were identified in this study. Of 8083 patients, 6531 had an aTL (81%) and 1552 patients had an iTL (19%). High risk features in the iTL group included cN1 (n=194), ETE (n=645), positive margins (n=567), M1 disease (n=25), pN1 (n=571), LVI (n=307), and poorly differentiated cancer (n=42). 12% (n=1001) and 5% (n=368) of the patients had more than one or more than two high-risk features respectively. The mean age (standard deviation) of patients in the aTL and iTL group was 47.4 (14.8) and 47.8 (15.9) years respectively (Table 1). The proportion of patients less than and more than 45 years of age was similar between the two groups (p=.691). The number of female compared to male patients was higher in each group (p=.0007). Compared to those in the aTL group, patients who had iTL had more comorbidities (16% vs 13%, p=.027), were more likely to be of White-Hispanic race (8% vs 6%, p<.0001), and have government insurance (27% vs 23%, p=.001). There was no difference in income status
8
(p=.777) and facility type (p=.064) between the two groups. Patients who had an iTL were more likely to have a tumor size between 2.1 to 4 cm (p=.026). On adjusted logistic regression analysis, the following factors were associated with iTL: presence of comorbidities (OR 1.26, 95% CI 1.96 – 1.50, p=.011), government insurance (OR 1.34, 95% CI 1.16 – 1.57, p<.007), no insurance (OR 1.53, 95% CI 1.11 – 2.10, p=.009), tumor size 2.1 to 4 cm (OR 1.26, 95% CI 1.10 – 1.45, p=.001) and White-Hispanic race (OR 1.58, 95% CI 1.20 – 2.09, p=.001).
Survival Analysis Mean overall survival in the iTL group was lower compared to that in the aTL group (9.81 years vs 10.75 years). Patients in the iTL group had significantly lower 5-year and 10-year overall survival as compared to those in the aTL group (94.7% vs 97.5% and 88.5% vs 92.4% respectively, log rank p<.001). On propensity score matching analysis, patients in the iTL group still had significantly lower 5-year and 10-year overall survival as compared to those in the aTL group (94.7% vs 96.5% and 88.5% vs 90.5% respectively, log-rank p=.004). (Table 2, Figure 1) On univariate Cox analysis, iTL was associated with poor survival as compared to aTL (HR 1.98, 95% CI 1.52 – 2.59, p<0.001). Other factors associated with adverse survival were age > 45 years (p<.001), male gender (p<.001), presence of comorbidities (p<.001), government insurance (p<0.001) and low-income status (p<.001). iTL group consisted of patients with only high-risk features, therefore these high-risk features were not included again as independent variables for adjusted survival analysis. On multivariable Cox survival analysis, overall survival was worse in patients with iTL (HR 1.97, 95% CI 1.51 – 2.58, p<.001), age > 45 years (HR 4.63, 95% CI 3.06 – 7.00, p<.001), male gender (HR 1.53, 95% CI 1.18 – 2.00, p<.001), presence of
9
comorbidities (HR 1.72, 95% CI 1.31 – 2.26, p<.001), government insurance (HR 4.88, 95% CI 3.71 – 6.43, p <.001) and low income status (HR 1.60, 95% CI 1.19 – 2.14, p=.002). (Table 3) In a separate subgroup analysis of patients without metastatic disease but other high-risk factors, iTL was still associated with a reduced 5-year (95.3% vs 97.5%) and 10-year (89.1% vs 92.4%) overall survival compared to patients undergoing aTL (log rank p<.001). A subsequent comparison between patients in iTL and TT+RAIT groups was performed. (Table 4) In comparison to iTL group (n=1552), patients in TT+RAIT group (n=36938) who had total thyroidectomy and RAI for similar tumors with high-risk features had significantly improved 5-year (97.1% vs 94.7%) and 10-year (92.3% vs 88.5%) overall survival (log rank p<.001). (Figure 2) Total thyroidectomy with subsequent RAI treatment was associated with improved survival on Cox univariate (HR .56, 95% CI .44 – .70, p<.001) and adjusted multivariable survival analysis (HR .65, 95% CI .51 – .83, p<.001). On propensity score matching analysis, patients in the TT+RAIT group maintained survival advantage as compared to those in the iTL group (5 years- 96.5% vs 94.7% and 10 years- 90.9% vs 88.5% respectively, log-rank p=.031). (Table 5)
Discussion We conducted a statistically well-powered retrospective study analyzing patients with T1b and T2 PTC in the NCDB treated from 2004 to 2014. These patients were treated before publication of the pivotal 2015 ATA guidelines where lobectomy was for the first time formally recommended as an acceptable alternative to total thyroidectomy for PTC with a tumor size 1.1 to 4 cm. However, there were 1552 patients had a lobectomy for these tumors when total thyroidectomy should have been the recommended operation. Applying the ATA guidelines, we
10
identified that approximately 19% of the patients had a iTL as the definitive operation for their T1b – T2 PTC with high-risk features. In theory these patients should have had an initial TT or a subsequent completion thyroidectomy. This may be due to inadequate preoperative evaluation leading to inappropriate patient selection, lack of review of the postoperative pathologic results identifying high-risk features or patients who refused to have a completion thyroidectomy or who were lost to follow up. The high incidence of thyroid lobectomy in patients with high-risk features suggests there is a need to improve and standardize the preoperative and postoperative evaluation of these patients and to develop protocols to ensure the appropriate operation is performed. A detailed imaging of thyroid nodule and cervical lymph nodes, availability of dedicated endocrine cytopathologist, assessment of vocal cord mobility, discussion in multidisciplinary tumor board, and a system to track and follow patients could be considered for such management protocols. Individual high-risk features have been shown to be associated with increased recurrence and reduced survival in PTC. In a study of 47,902 patients with PTC, Adam et al. concluded that presence of clinical lymph node metastasis was linked with poorer overall survival (HR 1.32 95% CI 1.04-1.67, p=.021) (6). Those authors advised preoperative screening for lymph node metastases in all patients with PTC. Many patients with PTC can have very long periods of progression-free survival even in the presence of distant metastases. This was shown by a single institutional study of 3664 patients from Memorial Sloan Kettering Cancer Center where a 77.6% five-year survival could be achieved (7). However, such excellent survival outcomes require complete extirpation of thyroid tissue with subsequent radioiodine therapy. In a study from the NCDB, Youngswirth et al. concluded that both minimal and extensive ETE were associated with reduced overall survival in patients with differentiated thyroid cancers (8). This
11
finding carries significance as extensive ETE could potentially be identified preoperatively as reported by Kuo et al. They showed that a detailed sonographic assessment of ETE led to a reduced rate of potentially avoidable total thyroidectomy from 57% to 31% (9). With this approach they were able to reliably exclude ETE preoperatively in most patients. Pathology specimens must be reviewed to identify any high-risk histological features when present. Pathologically positive neck lymph nodes are a major determinant in PTC staging and affect the need for subsequent RAI treatment and cancer surveillance of PTC. Therefore, detection of positive lymph nodes in the thyroid pathology specimens requires that a completion thyroidectomy be considered to enable RAI treatment and to decrease the recurrence rate (2). Although rare, poorly differentiated thyroid cancers (PTDC) are a major survival determinant of in nonanaplastic follicular cell derived thyroid cancers. Total thyroidectomy is the recommended surgical treatment for these patients (10). Ideally, PDTC should be identified on preoperative evaluation. When PDTC is found after thyroid lobectomy, a completion thyroidectomy is strongly advised. Similarly, microscopic and macroscopic positive margins have been found to be associated with reduced overall survival in patients with PTC (11). This is a high-risk feature for recurrence of PTC so a completion thyroidectomy should be considered. Additionally, LVI also affects survival independent of other high-risk features as described by Pontius et al. (12). Presence of LVI in the final pathology specimen ideally warrants a completion thyroidectomy to enable RAI treatment. RAI treatment is associated with improved survival in these patients (12). Patients with postoperatively detected adverse tumor features must be tracked and advised of the seriousness of these features to ensure an optimal oncologic outcome. While the ATA guidelines recommend either a total/near total thyroidectomy or a lobectomy for PTC 1 – 4 cm without extra thyroidal extension and without clinical evidence of
12
lymph node metastasis, much of the data supporting use of lobectomy for PTC has been derived from selective cancer centers. The 2015 ATA guidelines were primarily based on supporting evidence for unilateral thyroid surgery for T1-2 PTC from two large volume academic cancer centers from US and Japan (13, 14). These centers tend to have rigorous preoperative protocols that lead to better patient selection as well as postoperative surveillance protocols to identify patients that meet the criteria for completion thyroidectomy. Smaller community programs often lack a multi-disciplinary perioperative oncology team so the use of lobectomies in these centers may not achieve the favorable results of academic centers. Patients that underwent iTL had significantly poorer overall survival at 5 and 10 years after their operation than those who underwent aTL. This is an expected finding as patients in the iTL group had high-risk features that may not have been accounted for in the operative decisionmaking process and resulted in the poorer prognosis. Propensity score matching was done to ensure that similar patients were compared between both groups. The negative effect of iTL on overall 5 and 10-year survival was still observed. Even after adjusting for baseline confounding variables, patients who had aTL were twice as likely to survive as those who underwent iTL. Even though the number of patients with metastatic disease was low in our study due to a concern that it might disproportionately affect overall survival, we decided to exclude these patients and do a separate subgroup analysis. The result confirmed the main finding that overall survival was better for patients in aTL group as compared to the iTL group. Overall survival showed significant improvement when patients with > 1 to 4 cm PTC and high-risk features underwent total thyroidectomy with subsequent RAI. This finding was not in agreement with results of the 2014 study by Adam et al. (15) They found no survival difference between patients undergoing thyroid lobectomy and total thyroidectomy for T1b-T2 PTC. Selection of patients
13
with only high-risk features, and comparison between patients undergoing either inappropriate lobectomy or appropriate total thyroidectomy with RAI were the major differences in our study compared to that by Adam et al. Our results emphasize the need for a thorough preoperative assessment of risk factors for patients diagnosed with PTC. Once high-risk features are identified pre- or post-operatively, these patients should be advised that TT or completion thyroidectomy, respectively may provide a better outcome. To our knowledge this is the first study that has reported the high proportion of patients undergoing inappropriate lobectomies for PTC in the US and its negative effect on overall survival. We also used propensity score matching to eliminate the effect of confounders on our results. The findings of our study are important for thyroid surgeons to improve their patient outcomes. Results of our study can also be used to better inform patients with PTC when making their decision regarding the type of thyroidectomy. Our study does have several limitations. The NCDB does not have data on disease specific survival. However, PTC is a slow growing cancer with excellent long-term survival and we believe that overall survival is a reliable proxy for disease specific survival. There is also no information on patient preferences or intra- or post-operative decision making that could have altered the choice between TL and TT in these patients. Additionally, the NCDB lacks data about recurrence of PTC, hence, the cause of better survival remains presumptive.
Conclusions The results of our study show that a substantial fraction of patients with T1b-2 PTC and high-risk features underwent iTL instead of TT+RAIT. Our results also suggest that iTL has a significant negative effect on long-term survival of these patients. Surgeons performing TL and TT now
14
have an opportunity to evaluate their individual statistics. The focus should be on improving effectiveness of pre- and post-operative processes with a goal of minimizing the rate of iTL in an individual’s surgical practice. Thyroid surgeons should adhere to and carefully individualize the applicability of treatment guidelines when determining the type of thyroidectomy for their patients with PTC.
15
References 1. James BC, Timsina L, Graham R, et al. Changes in total thyroidectomy versus thyroid lobectomy for papillary thyroid cancer during the past 15 years. Surgery 2019;166(1):4147. 2. Haugen BR, Alexander EK, Bible KC, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: The American thyroid association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016;26(1):1-133. 3. Nickel B, Tan T, Cvejic E, et al. Health-related quality of life after diagnosis and treatment of differentiated thyroid cancer and association with type of surgical treatment. JAMA Otolaryngol Head Neck Surg. 2019 Jan 17. [Epub ahead of print] 4. Ullmann TM, Gray KD, Stefanova D, et al. The 2015 American thyroid association guidelines are associated with an increasing rate of hemithyroidectomy for thyroid cancer. Surgery 2019 May 2. [Epub ahead of print] 5. Vargas-Pinto S, Romero Arenas MA. Lobectomy compared to total thyroidectomy for low-risk papillary thyroid cancer: A systematic review. J Surg Res 2019;242:244-251. 6. Adam MA, Pura J, Goffredo P, et al. Presence and number of lymph node metastases are associated with compromised survival for patients younger than age 45 years with papillary thyroid cancer. J Clin Oncol 2015;33(21):2370-2375. 7. Wang LY, Palmer FL, Nixon IJ, et al. Multi-organ distant metastases confer worse disease-specific survival in differentiated thyroid cancer. Thyroid 2014;24(11):15941599.
16
8. Youngwirth LM, Adam MA, Scheri RP, et al. Extrathyroidal extension is associated with compromised survival in patients with thyroid cancer. Thyroid 2017;27(5):626-631. 9. Kuo EJ, Thi WJ, Zheng F, et al. Individualizing surgery in papillary thyroid carcinoma based on a detailed sonographic assessment of extrathyroidal extension. Thyroid 2017;27(12):1544-1549. 10. Ibrahimpasic T, Ghossein R, Carlson DL, et al. Outcomes in patients with poorly differentiated thyroid carcinoma. J Clin Endocrinol Metab 2014;99(4):1245-1252. 11. Youngwirth LM, Adam MA, Scheri RP, et al. Patients treated at low-volume centers have higher rates of incomplete resection and compromised outcomes: Analysis of 31,129 patients with papillary thyroid cancer. Ann Surg Oncol 2016;23(2):403-409. 12. Pontius LN, Youngwirth LM, Thomas SM, et al. Lymphovascular invasion is associated with survival for papillary thyroid cancer. Endocr Relat Cancer 2016;23(7):555-562. 13. Nixon IJ, Ganly I, Patel SG, et al. Thyroid lobectomy for treatment of well differentiated intrathyroid malignancy. Surgery 2012;151(4):571-579. 14. Matsuzu K, Sugino K, Masudo K, et al. Thyroid lobectomy for papillary thyroid cancer: long-term follow-up study of 1,088 cases. World J Surg 2014;38(1):68-79. 15. Adam MA, Pura J, Gu L, et al. Extent of surgery for papillary thyroid cancer is not associated with survival: an analysis of 61,775 patients. Ann Surg 2014;260(4):601-605.
17
Table 1. Comparison of Demographic and Clinical Characteristics of the Patients Having an Appropriate or an Inappropriate Thyroid Lobectomy Appropriate Inappropriate Variable All thyroid thyroid p Value lobectomy lobectomy Age Mean years ± 47.5 ± 15.1 47.4 ±14.8 47.8 ± 15.9 .108 SD ≤ 45 y, n (%) 3,612 (45) 2,911 (45) 701 (45) .691 > 45 y, n (%) 4,471 (55) 3,620 (55) 851 (55) Sex, n (%) Male 1,743 (22) 1,369 (21) 374 (24) .007* Female 6,340 (78) 5,162 (79) 1,178 (76) Comorbidity, n (%) No 6,971 (86) 5,660 (87) 1,311 (84) .027* Yes 1,112 (14) 871 (13) 241 (16) Race, n (%) White non5,884 (72) 4,756 (73) 1,088 (70) <.001* Hispanic White 495 (6) 369 (6) 126 (8) Hispanic African 656 (8) 559 (9) 97 (6) American Other 1,088 (14) 847 (12) 241 (16) Insurance, n (%) None 383 (5) 298 (5) 85 (5) .001* Private 5,775 (71) 4,725 (72) 1,050 (68) Government 1,925 (24) 1,508 (23) 417 (27) Income, n (%) Low 1,150 (14) 933 (14) 217 (14) .777 High 6,884 (85) 5,556 (86) 1,328 (86) Cancer center, n (%) Academic 2,037 (37) 1,618 (37) 419 (41) .020* Non3,399 (63) 2,788 (63) 611 (59) academic Tumor size, cm Mean size ± 2.13 ± 0.81 2.14 ± 0.8 2.18 ± 0.84 .026* SD 1.1 - 2, n (%) 4,554 (56) 3,719 (57) 835 (54) 2.1 - 4, n (%) 3,529 (44) 2,812 (43) 717 (46) *Significant.
18
Table 2. Survival Analysis for Papillary Thyroid Cancer Patients Having an Inappropriate Thyroid Lobectomy Compared to Those Having an Appropriate Thyroid Lobectomy Overall survival Patient group Mean ± SE, y
5-year, %
10-year, %
10.75 ± .03
97.5
92.4
p Value
All patients Appropriate lobectomy
<.001* Inappropriate lobectomy
9.81 ± .07
94.7
88.5
10.39 ± .06
96.5
90.5
Propensity score matched patients Appropriate lobectomy
.004* Inappropriate lobectomy *Significant.
9.81 ± .07
94.7
88.5
19
Table 3. Univariate and Multivariable Cox Regression Analysis for Factors Associated with Overall Survival Univariate Multivariable Predictor HR 95% CI p Value HR 95% CI p Value Lobectomy Appropriate 1 1 Inappropriate 1.98 1.52 - 2.59 <.001* 1.97 1.51 - 2.58 <.001* Age ≤ 45 y 1 1 > 45 y 7.56 5.04 - 11.34 <.001* 4.63 3.06 - 7.00 <.001* Sex Female 1 1 Male 1.79 1.38 - 2.32 <.001* 1.53 1.18 - 2.00 .001* Comorbidity No 1 1 Yes 3.02 2.32 - 3.19 <.001* 1.72 1.31 - 2.26 .001* Race White non-Hispanic 1 1 Other 0.37 0.22 - 0.63 <.001* 0.37 0.21 - 0.63 <.001* White Hispanic 0.47 0.23 - 0.95 .037* 0.44 0.22 - 0.90 .025* African American 1.27 0.85 - 1.89 .247 0.85 0.56 - 1.30 .446 Insurance Private 1 1 None 2.47 1.35 - 4.53 .004* 2.51 1.36 - 4.63 .003* Government 7.37 5.64 - 9.60 <.001* 4.88 3.71 - 6.43 <.001* Income High 1 1 Low 2.01 1.52 - 2.66 <.001* 1.60 1.19 - 2.14 .002* Cancer center Academic 1 Non-academic 0.99 0.76 - 1.29 .920 Tumor size, cm 1.1 - 2 1 2.1 - 4 1.17 0.91 - 1.49 .216 *Significant. HR, hazard ratio.
20
Table 4. Comparison of Demographic and Clinical Characteristics of the Patients Having an Inappropriate Thyroid Lobectomy or Total Thyroidectomy with Radioactive Iodine Treatment Variable All iTL TT+RAIT p Value Age Mean size ± SD 45.9 ± 14.9 48.1 ±16.1 45.8 ± 14.9 <.001* ≤ 45 y, n (%) 18,670 (48) 701 (45) 17,969 (49) .004* > 45 y, n (%) 19,820 (52) 851 (55) 18,969 (51) Sex, n (%) Male 9,977 (26) 374 (24) 9,603 (26) .050* Female 28,513 (72) 1,178 (76) 27,335 (74) Comorbidity, n (%) No 33,096 (86) 1,311 (84) 31,785 (86) .043* Yes 5,394 (14) 241 (16) 5,153 (14) Race, n (%) White non-Hispanic 27,084 (70) 1,088 (70) 25,996 (65) <.001* White Hispanic 4,023 (10) 126 (8) 3,897 (11) African American 1,651 (4) 97 (6) 1,554 (4) Other 5,732 (16) 241 (16) 5,491 (20) Insurance, n (%) None 1,644 (4) 85 (5) 1559 (4) <.001* Private 29,028 (75) 1,050 (72) 27,978 (76) Government 7,818 (21) 417 (23) 7,401 (20) Income, n (%) Low 4,380 (11) 217 (14) 4,163 (11) .001* High 33,845 (89) 1,328 (86) 32,517 (89) Cancer center, n (%) Academic 10,974 (45) 419 (41) 10,555 (45) .003* Non-academic 13,488 (55) 611 (59) 12,877 (55) Tumor size, cm Mean size ± SD 2.15 ± 0.8 2.18 ± 0.84 2.15 ± 0.8 .329 1.1 - 2, n (%) 20,932 (54) 835 (54) 20,097 (54) 2.1 - 4, n (%) 17,558 (46) 717 (46) 16,841 (46) *Significant. iTL, inappropriate thyroid lobectomy; TT+RAIT, total thyroidectomy with radioactive iodine treatment.
21
Table 5. Survival Analysis for Papillary Thyroid Cancer Patients Having an Inappropriate Thyroid Lobectomy Compared to Those Having a Total Thyroidectomy with Radioactive Iodine Treatment Overall survival Patient group
5-year, %
10-year, %
p Value
Hazard ratio
95% CI
p Value
Inappropriate lobectomy
94.7
88.5
<.001*
1
Multivariable
<.001*
Total thyroidectomy
97.1
92.3
.65
.51 - .83
94.7
88.5
96.5
90.9
All patients
Propensity score matched patients Inappropriate lobectomy Total thyroidectomy *Significant.
.031*
22
Figure Legends Figure 1. Kaplan-Meier plot comparing survival between appropriate lobectomy (aTL) and inappropriate lobectomy (iTL) groups. Figure 2. Kaplan-Meier plot comparing survival between inappropriate lobectomy (iTL) and total thyroidectomy with radioactive iodine treatment (TT+RAIT) groups.
23
Precis A total of 19.2% patients with T1b-T2 (1.1 – 4 cm in size) papillary thyroid cancer had high-risk features and underwent inappropriate thyroid lobectomy with an adverse impact on overall survival.