Recurrent Middle Eastern Differentiated Thyroid Carcinoma Has Worse Outcomes Than Persistent Disease

Background: Despite the excellent prognosis of differentiated thyroid carcinoma (DTC), recurrent and persistent disease remain major challenges. Emerging studies to differentiate between recurrent and persistent disease are controversial, with studies from the Middle East lacking. Methods: We retrospectively analyzed 1691 patients who underwent surgery ± I131 treatment for DTC, with a median age of 38.7 years and median follow-up of 95.3 months. Results: We found a similar prevalence rate for persistent and recurrent disease (17.7% vs. 17.9%) in Middle Eastern DTC patients. Relative to patients with persistent disease, patients with recurrent disease were significantly older (median age: 36.1 vs. 45.8 years; p < 0.0001) and were more likely to have ATA high-risk tumors (61.5% vs. 75.2%; p = 0.0003). On multivariate logistic regression analysis, both T and N status were independent predictors for recurrent as well as structural persistent disease. However, older age, bilaterality and extrathyroidal extension were independent predictors of recurrent disease alone. In addition, patients with recurrent disease had significantly worse cancer-specific survival (p < 0.0001), which remained significant in multivariate analysis. Conclusions: Although persistent and recurrent disease in Middle Eastern DTC have similar frequencies, recurrent disease has worse outcomes compared to persistent disease. Hence, differentiating recurrence from persistence has great potential clinical relevance for therapeutic and follow-up approaches, contributing to improving the outcomes of DTC patients of Middle Eastern ethnicity.


Introduction
Differentiated thyroid carcinoma (DTC) is the commonest endocrine malignancy and generally associated with excellent survival [1][2][3].Adequate initial management, including surgery, in some cases followed by radioactive iodine (RAI) and thyrotropin suppression therapy, leads to a high likelihood of cure in these patients [4][5][6][7].Despite the high cure rate, disease recurrence and persistence remain major challenges for patients and physicians [8,9].
Most published studies have considered persistent and recurrent DTC to form a single entity [10][11][12].The definition and timeline of recurrence and persistence is universal in the literature.According to the recent ATA guidelines, disease-free status is defined as no clinical evidence of a tumor, no evidence of a tumor based on RAI imaging or ultrasound, unstimulated Tg < 0.2 mg/mL or stimulated Tg < 1 mg/mL in the absence of antibodies.
However, from a clinical perspective, the disease is considered persistent when detected a short time after initial therapy and recurrent when detected after a disease-free period of ≥one year [4].Although a few studies have tried to differentiate between persistent and recurrent disease [13,14], they could not identify any differences between these two conditions with regard to patient prognosis.However, one recent study involving a large series of DTC patients of European ethnicity has demonstrated a significant difference between persistent and recurrent disease in terms of outcomes and clinical implications [15].
Previously, we investigated the recurrence rate in Middle Eastern papillary thyroid carcinoma (PTC) [16].However, we did not investigate the difference between recurrence and persistence in DTC in the Middle East, where PTC is considered the second commonest cancer affecting females [17].Most studies have considered recurrence and persistence to be a single entity or used them interchangeably.Their impact on management and outcomes for these two entities separately has not been explored thoroughly.There is ongoing debate about the necessity to differentiate between persistent and recurrent disease, and the results have been inconsistent.Moreover, there is a paucity of data regarding recurrence and persistence in patients of this ethnicity.A better understanding of these entities could result in better patient management and surveillance.
The goal of this study was to determine the frequency of recurrent and persistent diseases, their clinico-pathological associations and, most importantly, the difference in outcomes (cancer-specific survival) between recurrent and persistent disease in DTC, as well as its clinical implications.To achieve this, we retrospectively evaluated a series of 1691 Middle Eastern patients with DTC.

Patient Selection and Clinico-Pathological Data
One thousand eight-hundred and twenty-two consecutive DTC patients diagnosed between 1988 and 2020 at King Faisal Specialist Hospital and Research Centre (Riyadh, Saudi Arabia) were eligible to be included in this study.Patients of Middle Eastern origin were included in the study, with patients of other nationalities/ethnicities excluded.Also, patients with follow-up durations of less than 5 years who had not developed recurrent disease were excluded.Finally, 1691 patients were included for analysis.Cases were identified based on clinical history followed by fine-needle aspiration cytology for confirmation.The Institutional Review Board of the hospital approved this study, and since only retrospective patient data were utilized, the Research Advisory Council (RAC) provided a waiver of consent under project RAC # 221 1168 and # 2110 031.
All patients underwent surgery (81.5% underwent total thyroidectomy and 18.5% underwent hemithyroidectomy/lobectomy), performed by board-certified, highly experienced endocrine surgeons who have performed high volumes of thyroidectomies (>25 thyroidectomies/year).Given that the patients included in our study had been diagnosed between 1988 to 2020, the management protocols varied based on the changes in international standards over time.However, our institute has been following the recommendations of the American Thyroid Association (ATA) guidelines since they were first published in 1996 and modified the protocols based on subsequent revisions of the ATA guidelines.
Overall, 77.9% (1317/1691) of patients received RAI therapy.Patients were seen 6 to 8 weeks following surgery, having been prepared with thyroid hormone withdrawal for at least four weeks and a low-iodine diet for one week.The ATA guidelines [4] were used to determine the RAI dosages administered.Previously, the administered dosage used to be 100 mCi for remnant ablation, 100-150 mCi if there are lymph node metastases and 150-200 mCi if there are distant metastases.However, in the last 5 years, a trend towards more conservative doses of RAI was adopted, with most patients receiving 30-100 mCi for thyroid remnant ablation and 100-200 mCi for patients with lymph node or distant metastases.In addition, for pediatric patients, the dosage was adjusted to the child's body weight (based on adult dosage of 70 kg person).
Baseline clinico-pathological data were collected from case records and are summarized in Table 1.Staging of DTC was performed using the eighth edition of American Joint Committee on Cancer (AJCC) staging system [18].

Definitions of Persistent and Recurrent Disease
All patients were followed up at regular intervals after surgery (±RAI therapy) with neck ultrasound, CT scan, post-therapy 131I WBS (for patients who underwent RAI therapy), a thyroid function test and serum Tg and anti-Tg antibody measurements (Tg and anti-Tg were analyzed only for patients who underwent total thyroidectomy).
Patients were grouped according to disease status, with patients considered as having an excellent response if there was an absence of clinical, biochemical (unstimulated serum thyroglobulin (Tg) levels of <0.2 µg/L or stimulated Tg levels of <1 µg/L in the absence of interfering thyroglobulin antibodies (TgAb)) or structural disease (locoregional lymph node metastasis as evidenced by neck ultrasound and pathologically confirmed by fine needle aspiration biopsy or the appearance of distant metastases in the bone, lung or brain as evidenced by 131I WBS or CT scan).In contrast, active disease was defined by the presence of biochemical (unstimulated serum Tg levels ≥ 0.2 µg/L or stimulated Tg levels ≥ 1 µg/L; the rising or de novo appearance of TgAb) or structural (abnormal findings on radio-imaging) disease.Presence of active disease within 12 months post-operatively was defined as persistent disease, whereas patients who had excellent response for at least 12 months post-operatively and later developed active disease were defined as having recurrent disease, based on ATA guidelines and previous publications [4,13,15].

Study End-Point
The study end-point was marked by an analysis of the difference in outcomes (cancerspecific survival (CSS)) between recurrent and persistent disease in DTC.CSS was defined as the time from diagnosis to death due to DTC progression.

Statistical Analysis
The associations between clinico-pathological variables and recurrence/persistence were determined using contingency table analysis and Chi square tests.A Mantel-Cox log-rank test was used to evaluate CSS.Survival curves were generated using the Kaplan-Meier method.Logistic regression analysis and a Cox proportional hazards model was used for univariate and multivariate analysis.Two-sided tests were used for statistical analyses, with the limit of significance defined as a p value < 0.05.Data analyses were performed using the JMP14.0(SAS Institute, Inc., Cary, NC, USA) software package.

Incidence and Treatment of Persistent and Recurrent Disease
The majority of the patients with DTC (64.4%; 1089/1691) had an excellent response after initial surgery ± RAI therapy.Persistent disease was noted in 17.7% of patients (299/1691), with biochemical persistence accounting for 27.8% (83/299) and structural persistence for 72.2% (216/299) of these cases.Recurrent disease was seen in 17.9% (303/1691) of patients in our cohort, with biochemical recurrence accounting for 26.7% (81/303) and structural persistence for 73.3% (222/303) of these cases (Table 2).Supplementary Figures S1 and S2 show the initial treatments performed, the outcome after the initial treatment (in terms of recurrence and persistence after 12 months) and additional treatments that were performed after recurrent/persistent disease.

Survival Outcomes for Persistent and Recurrent Disease
Patients with recurrent disease had a significantly worse CSS (p < 0.0001) compared to those with persistent disease (Figure 1A).We further divided the persistent and recurrent disease patients into structural and biochemical subgroups given the different risk profiles and treatment regimens for these two subgroups.Patients with structural recurrent disease had a significantly worse CSS (p < 0.0001) compared to those with structural persistent disease (Figure 1B).On multivariate Cox proportional hazards analysis, recurrent disease was found to be an independent predictor of worse CSS (hazard ratio = 6.44; 95% confidence interval = 2.40-22.56;p < 0.0001) (Table 5).However, the difference in CSS was not significant between patients with biochemical recurrent and biochemical persistent disease (p = 0.2670) (Figure 1C).

Long-Term Outcomes for DTC Patients with Persistent Disease
Of the 299 patients with persistent disease, 161 (53.8%) patients underwent additional surgery alone and 22 (7.4%) had additional surgery and RAI therapy, whereas 95 (31.8%) patients received additional doses of RAI therapy alone (ranging from two to seven).Twenty-one patients did not receive any additional therapies (Supplementary Figure S2).Following additional therapies, 142 (48.8%) of the 299 patients with persistent disease were free of disease.At long-term follow-up, 16 of these patients had a relapse and 126 patients remained disease-free at the last follow-up (Table 2).

Discussion
In this study, we retrospectively analyzed a large cohort of 1691 DTC patients of Middle Eastern ethnicity.Overall, recurrence after initial treatment is relatively common and has been reported to vary between 3% and 23% [19][20][21][22], whereas persistent disease has been reported in in up to 23% of DTCs [15,23,24].We found that 35.6% (602/1691) of DTC patients had disease events (persistence or recurrence) after initial therapy.This had a huge impact on the patients' subsequent therapy and follow-up.
With a relatively high rate of disease events, we ended up with a large investigated cohort of 602 cases with disease after initial therapy, which adds strength to our findings.Interestingly, in this cohort, we found a nearly equal frequency of recurrent (50.3%) and persistent disease (49.7%).Most of the previous studies [10][11][12] have evaluated and considered recurrent and persistent disease as a single entity despite the differentiation being made recently by the ATA guidelines, where clear definitions of both conditions have been clarified [4].However, the distinction between persistent and recurrent disease is still controversial in the literature [13][14][15] and is lacking in DTCs from Saudi Arabia, where PTC (the commonest type of DTC) ranks as the second most common cancer affecting Saudi females, after breast cancer.
Although no significant difference in incidence between recurrent and persistent disease (16.6% vs. 16.4%) was noted, we found that T3/T4 status and cervical lymph node involvement (N1a/b) were likely to predict persistent DTC.Lymph node metastasis was also a predictor of recurrent disease.However, older age, bilateral tumors and ex-trathyroidal extension were independent predictors of recurrent disease, but not persistent disease.These variables have been shown to increase the risk of recurrence in previous studies [13,[25][26][27].
In this study, DTC recurrence and persistence were, in fact, considered two separate clinical conditions in terms of patient prognosis and outcomes.Despite equal frequencies of both conditions, patients with recurrent disease clearly had worse CSS, which remained significant on multivariate analysis.The worse outcome of recurrent disease is most likely the consequence of more aggressive clinical associations at presentation.Furthermore, the incidence of ATA high-risk tumors in our cohort was relatively high.This could partially be attributed to the fact that, as a tertiary care institute, patients with advanced disease from all over Saudi Arabia are referred to our hospital.In addition, previous studies from the Middle East have shown a higher incidence of advanced disease [28,29], suggesting that either DTC in this population is inherently more aggressive or there is a delay in patients seeking medical help (or a combination of both).Additionally, rescue therapies with newer agents, such as Lenvatinib and Sorafenib, have only been used for patients with tumor recurrence at our institute since 2019.However, considering that the patients included in our cohort were diagnosed between 1988 and 2020, the majority of patients with recurrent tumors have not received these newer therapies.This could be another reason for the relatively high mortality rate in patients with tumor recurrence in our cohort.Indeed, on further classification, the worst outcome was limited to patients with structural recurrence, not those with biochemical recurrence.This has important clinical implications given the different risk profiles and treatment regimens for these two subgroups.
A previous large study [15] using the same clinical criteria for defining recurrence and persistence, conducted on a European cohort, was able to demonstrate differences in clinical outcomes between recurrent and persistent DTC.However, our results are opposite to their results, as they showed that persistent disease had worse outcomes than recurrent DTC.These contrasting results could be attributed to ethnic differences and might suggest differences in the severity of DTC at the time of presentation.Previous reports have also highlighted the role of ethnicity in the patient characteristics and clinical presentation of DTC among different ethnic groups [30][31][32].The association of metastasis with recurrence in Middle Eastern DTC may be indicative of a greater extent of disease at the time of presentation, placing patients at higher risk of true recurrence and worse outcomes.Indeed, a higher percentage of deaths was noted in patients with recurrent disease in our cohort due to the development of distant metastasis and progression of disease.Of the 303 patients with recurrent disease, 11.2% (34/303) died overall, and 32 of these 34 patients died due to development of distant metastasis, whereas of the 299 patients with persistent disease, 2.3% (7/299) died overall, and 3 of these 7 patients died due to development of distant metastasis.
One of the limitations of this study is that it is retrospective and from a single tertiary care institute, which could give rise to selection bias.Also, the findings of this study were obtained from a very specific ethnicity, and therefore, generalization of the conclusions should be undertaken with caution.Additionally, although the surgeries were performed by board-certified, highly experienced endocrine surgeons who have performed high volumes of thyroidectomies, we do acknowledge that some patients might have had inadequate initial treatment, leading to increased incidence of recurrent/persistent disease.

Conclusions
This study showed that Middle Eastern patients with DTC after initial therapy had similar frequencies of persistent and recurrent disease.However, differences in predictive risk factors in patients with recurrent and persistent disease were observed.Recurrent disease was found to be associated with more aggressive clinico-pathological characteristics and was a strong predictor of worse cancer-specific survival compared to persistent disease in our cohort of Middle Eastern DTC patients.Our findings will help physicians and surgeons in understanding the difference between persistent and recurrent disease and -quartile range, ATA-American Thyroid Association, RAI-radioactive iodine, mCi-millicurie, S.D.-standard deviation.

Figure 1 .
Figure 1.Cancer-specific survival.(A) Kaplan-Meier survival curve showing poor cancer-s survival in patients with recurrent disease compared to those with persistent disease.(B) Ka Meier survival curve showing poor cancer-specific survival in patients with structural rec disease compared to those with structural persistent disease.(C) Kaplan-Meier survival showing no significant difference in cancer-specific survival in patients with biochemical rec disease compared to those with biochemical persistent disease.

Figure 1 .
Figure 1.Cancer-specific survival.(A) Kaplan-Meier survival curve showing poor cancer-specific survival in patients with recurrent disease compared to those with persistent disease.(B) Kaplan-Meier survival curve showing poor cancer-specific survival in patients with structural recurrent disease compared to those with structural persistent disease.(C) Kaplan-Meier survival curve showing no significant difference in cancer-specific survival in patients with biochemical recurrent disease compared to those with biochemical persistent disease.

Table 1 .
Clinico-pathological characteristics of differentiated thyroid carcinoma patient cohort.

Table 2 .
Incidence of persistent and recurrent disease in patients with DTC.

Table 3 .
Clinico-pathological associations of persistent and recurrent disease compared to excellent response in patients with differentiated thyroid carcinoma.

Table 4 .
Multivariate analysis of risk factors for predicting persistent and recurrent disease.

Table 5 .
Multivariate analysis of risk factors for predicting cancer-specific survival.

Table 5 .
Multivariate analysis of risk factors for predicting cancer-specific survival.