Abstract
KEY POINTS
Active surveillance is a feasible management strategy for carefully selected patients with locoregional recurrent differentiated thyroid cancer, particularly for small and stable lesions that do not threaten vital structures.
The overall progression rate of locoregional recurrence during active surveillance is approximately 23%, although this may be underestimated because the available studies are retrospective, highly heterogeneous, and affected by substantial loss to follow-up.
Patients with advanced initial AJCC stage and elevated baseline serum thyroglobulin levels appear to have a higher risk of progression.
SUMMARY
Background
Differentiated thyroid cancer (DTC) shows recurrence in fewer than 30% of cases. 1 When recurrence occurs, repeat surgery carries surgical risks and may be ineffective, particularly after multiple procedures.2,3 Therefore, in selected patients with small and stable recurrences, noninterventional strategies such as active surveillance have been adopted as alternatives.4–7 The most recent American Thyroid Association (ATA) guidelines suggest that suspicious lymph nodes or lesions measuring <8–10 mm in the shortest dimension may be followed without fine-needle aspiration (FNA) unless they grow or threaten vital structures. 8 Published studies and a recent meta-analysis reported a pooled progression rate of 28% for DTC. 9 However, because all available studies are retrospective, and some include small sample sizes and limited follow-up, important methodological limitations remain. The present study performed a systematic review and meta-analysis adjusted for loss to follow-up. 10
Methods
The systematic review and meta-analysis were performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The authors conducted a comprehensive literature search of MEDLINE, Embase, Scopus, and the Cochrane databases from inception to July 2025. Eligible studies included patients who underwent surgical removal of the primary thyroid cancer, had recurrent thyroid cancer (thyroid bed nodules or lymph node metastases), and provided sufficient data to estimate progression during active surveillance. The primary outcome was the progression rate of recurrent thyroid cancer during active surveillance, while the secondary outcome was the identification of factors associated with progression. Heterogeneity was assessed using Higgins inconsistency index (I2). A variable α was introduced to represent the proportion of patients lost to follow-up who experienced progression. Data extraction and quality assessment were performed independently by two authors using the Risk of Bias Assessment Tool for Nonrandomized Studies.
Results
Ten retrospective studies including 841 patients met the inclusion criteria, with study sample sizes ranging from 25 to 191 patients. The pooled mean follow-up was 51.6 months (95% confidence interval [CI], 36–67.2; I2, 99.1%). All studies included patients with DTC, predominantly papillary thyroid carcinoma. Diagnostic confirmation varied across studies: some reported pathological confirmation (FNA, washout thyroglobulin, or surgical confirmation), whereas others used imaging-based assessments alone or in combination with pathology (ultrasound, computed tomography, magnetic resonance imaging, or positron-emission tomography). Definitions of progression also differed among studies. Most defined progression as an increase of ≥3 mm, while others used criteria such as growth >2 mm, growth ≥ 3 mm/year, development of new metastases, or Response Evaluation Criteria in Solid Tumors (RECIST).
The overall pooled progression rate was 23% (95% CI, 12–34; I2, 93.3%). According to the diagnostic method, the pooled progression rate was 14% (95% CI, 0%–4%; I2, 93.4%) in imaging-based studies, 23% (95% CI, 12–34; I2, 93.3%) in studies using pathological confirmation or imaging, and 32% (95% CI, 15–55; I2, 83.2%) in biopsy-only studies. The pooled progression rate was significantly higher in studies with a follow-up duration <40 months compared with those with follow-up ≥40 months (36% [95% CI, 23–53; I2, 77.2%] vs. 17% [95% CI, 11–24; I2, 68.5%], p < 0.05). Patients in the progression group had significantly higher pooled log-transformed baseline serum thyroglobulin levels than those in the stable group (1.02 vs. −0.07, p < 0.05). Subgroup analysis also demonstrated a significantly higher progression rate among patients with a higher initial AJCC stage. After adjustment for loss to follow-up, progression rates increased from 20% to 24% to 61% at 42 months and from 9% to 24% to 85% at 84 months.
Conclusions
Active surveillance for locoregional recurrent DTC should be considered with caution, as reported progression rates may be underestimated because of substantial study heterogeneity and loss to follow-up. Higher progression rates have been reported in patients with advanced tumor stage and elevated baseline serum thyroglobulin levels.
COMMENTARY
This systematic review and meta-analysis reported substantial heterogeneity among studies evaluating active surveillance for DTC and suggested that the true progression rate may be underestimated. After adjustment for loss to follow-up, the pooled progression rate was estimated to increase to 35%–70%, and a higher risk of progression was observed in patients with a more advanced initial tumor stage and elevated baseline serum thyroglobulin levels. 10 Moreover, studies evaluating active surveillance (and consequently this meta-analysis) generally excluded patients with aggressive histology, distant metastases, PET-positive disease, or rapidly progressive lesions.
However, in everyday clinical practice, suspicious cervical lesions are frequently identified and safely monitored with serial ultrasound examinations over several years without evidence of significant structural progression, and this approach has already been widely implemented. Active surveillance is also supported by the latest ATA guidelines, which allow continued surveillance without FNA for suspicious lymph nodes or lesions measuring <8–10 mm in the shortest dimension. 8 The main challenge is to distinguish suspicious lymph nodes or lesions that are likely to progress from those that will remain stable over time. Differences in progression rates have been reported between studies including pathologically confirmed lesions and those based primarily on imaging diagnosis (32% vs. 14%). This discrepancy may reflect the fact that larger or clinically more aggressive lesions are more likely to undergo biopsy. In routine clinical practice, small lesions that are not threatening and are not adjacent to vital structures are often not biopsied if the result would not alter management. Therefore, the progression rates reported in studies based on imaging alone, or on combined imaging and pathological assessment, are likely more representative of real-world practice. The key issue is the identification of indolent lesions suitable for surveillance. When patients are appropriately selected, active surveillance appears to be a feasible and reasonable management strategy, particularly considering the well-recognized risks associated with repeated neck surgery, including nerve injury and permanent hypoparathyroidism, as well as the often limited efficacy of repeated surgical interventions.2,3
This meta-analysis also emphasized the overall low quality of evidence, as the currently available literature is limited to a small number of retrospective studies. Prospective studies are clearly needed. In addition, future research should aim to better identify which lesions are truly at higher risk and therefore require closer monitoring, additional diagnostic evaluation, or earlier intervention, as opposed to lesions that may be safely managed with long-term surveillance alone.