Introduction
Papillary thyroid carcinoma is the most common form of thyroid malignancy, accounting for 60–70 per cent of all thyroid cancer, and occurring more often in women in the fourth and fifth decades of life.Reference Shah, Loree, Dharker, Strong, Begg and Vlamis 1 Its incidence has more than doubled in the last three decades.Reference Davies and Welch 2 The mortality rate, however, has remained stable with a good prognosis, with a 10-year disease-related survival rate above 90 per cent.Reference Shah, Loree, Dharker, Strong, Begg and Vlamis 1 – Reference Hundahl, Fleming, Fremgen and Menck 3
Cervical lymph node metastases are found in 20–50 per cent of cases and in up to 90 per cent when micro-metastases are included. Cervical lymph node metastases are considered an independent risk factor for loco-regional recurrence and increased morbidity.Reference Qubain, Nakano, Baba, Takao and Aikou 4 – Reference Beasley, Lee, Eski, Walfish, Witterick and Freeman 8 Although cervical metastases are clearly associated with an increased recurrence rate, their effect on survival remains controversial.Reference Qubain, Nakano, Baba, Takao and Aikou 4 – Reference Beasley, Lee, Eski, Walfish, Witterick and Freeman 8 Mazzaferri and Young found that cervical metastases did not influence survival.Reference Mazzaferri and Young 9 However, other population-based studies demonstrated a negative effect on survival, with varying levels of significance.Reference Hundahl, Fleming, Fremgen and Menck 3 , Reference Zaydfudim, Feurer, Griffin and Phay 10 – Reference Ricarte-Filho, Ganly, Rivera, Katabi, Fu and Shaha 13
The current 2016 American Joint Committee on Cancer tumour–node–metastasis (TNM) staging system for papillary thyroid carcinoma incorporates only the presence and laterality of regional lymph nodes, yet prognosis is affected only in patients older than 45 years. Moreover, it was originally developed to predict risk for mortality rather than recurrence. Although the prognosis of papillary thyroid carcinoma patients with regional disease is quite variable, the American Joint Committee on Cancer TNM staging system does not take into consideration the burden of the metastatic regional disease, and does not differentiate between minimal or extensive metastatic disease in relation to effects on prognosis.
Lymph node ratio is defined as the number of metastatic lymph nodes divided by the number of lymph nodes examined. Lymph node ratio has been shown to be a prognostic variable in non-thyroidal cancers such as head and neck, gastric, breast, and colorectal cancers.Reference Ceelen, Van Nieuwenhove and Pattyn 14 – Reference Slidell, Chang, Cameron, Wolfgang, Herman and Schulick 16 In contrast, Roberts et al. demonstrated superior prognostic value for the number of positive lymph nodes, in comparison with lymph node ratio and American Joint Committee on Cancer nodal (N) staging for head and neck cancers.Reference Roberts, Colevas, Hara, Holsinger, Oakley-Girvan and Divi 17 Multiple recent studies have evaluated the prognostic value of lymph node ratio in papillary thyroid carcinoma regarding loco-regional recurrence and survival, with varying results.Reference Schneider, Chen and Sippel 18 – Reference Lee, Roh, Gong, Cho, Choi and Nam 26
This study aimed to review recent publications addressing the prognostic value of lymph node ratio in metastatic papillary thyroid carcinoma.
Materials and methods
A systematic literature review of the PubMed database was conducted on May 2016 using the search terms ‘papillary thyroid carcinoma’ and ‘lymph node ratio’. In addition, we manually searched the references of the identified articles and prior relevant reviews to detect further studies.
All articles that fulfilled our inclusion criteria were included. The studies had to be randomised or non-randomised controlled trials, or prospective or retrospective studies. The participants were post-thyroidectomy and neck dissection papillary thyroid carcinoma patients. There needed to be data regarding central or lateral cervical lymph nodes and lymph node ratio. Positive lymph nodes were required to be confirmed by pathological examination. Sufficient reported data were required for estimating the prognostic value of lymph node ratio on recurrence and survival.
Where available, the following data were extracted for analysis: number of patients, mean number of lymph nodes retrieved, mean number of metastatic lymph nodes, lymph node ratio, lymph node ratio threshold, hazard ratio, prognosis and survival.
This study was approved by the institutional review board of the Chaim Sheba Medical Center, Ramat-Gan, Israel (reference number: 3142-16-SMC).
Results
The electronic search primarily yielded 189 articles. Of these, 164 were excluded on the basis of the title or abstract, resulting in 25 potentially eligible studies. After critical examination, nine studies were considered adequate according to the inclusion criteria.Reference Schneider, Chen and Sippel 18 – Reference Lee, Roh, Gong, Cho, Choi and Nam 26
These nine studies were published between 2012 and 2016. All studies were retrospective. Overall, 12 400 papillary thyroid carcinoma patients were included. All patients had undergone thyroidectomy, and: prophylactic or therapeutic neck dissection; central or lateral compartment neck dissection; or both. Median age at the time of diagnosis was 48.6 years, and 76 per cent of patients were female.
One study utilised the Surveillance, Epidemiology, and End Results database of the National Cancer Institute to evaluate the impact of lymph node ratio in metastatic papillary thyroid carcinoma on disease-related mortality.Reference Schneider, Chen and Sippel 18 Eight studies evaluated the prognostic value of lymph node ratio on recurrence-free survival. The patients’ characteristics are summarised in Table I. A significant association was found between lymph node ratio and worse recurrence-free survival in 60 per cent and 75 per cent of the studies that investigated the effect of central compartment metastases and both central and lateral compartment metastases on recurrence-free survival, respectively.Reference Jeon, Yoon, Han, Yim, Hong and Song 19 – Reference Lee, Roh, Gong, Cho, Choi and Nam 26 Different lymph node ratio thresholds were used in each study (range, 0.3–0.86). The prognostic values of lymph node ratio and the hazard ratios for loco-regional recurrence are presented in Table II.
Table I Characteristics of patients in identified studies
N/A = not available; LNR = lymph node ratio
Table II Lymph node ratios and hazard ratios for recurrence-free survival in identified studies
LNR = lymph node ratio; N/A = not available; RFS = recurrence-free survival
Discussion
Cervical lymph node metastases are found in 20–50 per cent of papillary thyroid carcinoma patients and in up to 90 per cent when including micro-metastases. The current American Joint Committee on Cancer TNM staging system for papillary thyroid carcinoma incorporates only the presence and laterality of regional lymph nodes, and does not take into consideration the burden of metastatic regional disease. Lymph node ratio may have greater prognostic value, as it takes into consideration both the extent of surgery and the regional metastatic burden. Thus, it can differentiate between minimal and extensive metastatic disease, an issue missing from the American Joint Committee on Cancer TNM classification.
Although the data on the extent of nodal involvement and its subsequent effect on disease recurrence and overall survival rates are still debated, previous studies have demonstrated that cervical lymph node metastasis is correlated with loco-regional papillary thyroid carcinoma recurrence.Reference Qubain, Nakano, Baba, Takao and Aikou 4 – Reference Beasley, Lee, Eski, Walfish, Witterick and Freeman 8 Even though the lymph node ratio has prognostic significance in several non-thyroidal cancers, its significance in papillary thyroid carcinoma remains unclear.Reference Ceelen, Van Nieuwenhove and Pattyn 14 – Reference Roberts, Colevas, Hara, Holsinger, Oakley-Girvan and Divi 17 This study focused on the regional lymph node ratio in metastatic papillary thyroid carcinoma, and considered its prognostic significance on loco-regional recurrence and survival.
Only one study investigated the effect of lymph node ratio on survival. Schneider et al. utilised a population-level database (the Surveillance, Epidemiology, and End Results database) of the National Cancer Institute, and identified 10 955 metastatic papillary thyroid carcinoma patients treated with thyroidectomy and neck dissection.Reference Schneider, Chen and Sippel 18 Those authors demonstrated a significant association between lymph node ratio and disease-related mortality, with a hazard ratio of 4.33 (95 per cent confidence interval (CI) = 1.68–11.18, p < 0.01); there was a significant increase in disease-related mortality from 0.65 per cent for patients with a lymph node ratio of less than 0.42 to 1.72 per cent for patients with a lymph node ratio of 0.42 or more (p < 0.01). Yet, this association lost its significance when the analyses were performed after excluding patients with N1b disease, raising the question of whether the significance of lymph node ratio on survival may be attributed to lateral neck disease alone.
Central compartment analysis
Five out of eight studiesReference Jeon, Yoon, Han, Yim, Hong and Song 19 – Reference Ryu, Song, Choi, Roh, Nam and Kim 22 , Reference Vas Nunes, Clark, Gao, Chua, Campbell and Niles 24 investigated the effect of central compartment metastases (N1a) alone on recurrence. Three of the studiesReference Jeon, Yoon, Han, Yim, Hong and Song 19 , Reference Schneider, Mazeh, Chen and Sippel 21 , Reference Ryu, Song, Choi, Roh, Nam and Kim 22 (60 per cent) found a significant association between lymph node ratio and recurrence-free survival on multivariate analysis.
Jeon et al. investigated 292 papillary thyroid carcinoma patients who underwent prophylactic (63 per cent) or therapeutic (37 per cent) central neck dissection.Reference Jeon, Yoon, Han, Yim, Hong and Song 19 The median numbers of retrieved and metastatic lymph nodes were 9 (range, 6–13) and 3 (range, 2–5.5), respectively. The hazard ratio for recurrence in their study population increased significantly from 2.92 (95 per cent CI = 0.88–9.76, p = 0.081) in patients with a lymph node ratio of less than 0.4 to 14.83 (95 per cent CI = 5.07–43.35, p < 0.001) in patients with a lymph node ratio of 0.4 or more. They also found that lymph node ratio had a more distinct effect on prognosis as compared to lymph node size.
Lang et al. conducted their study on 51 clinically staged N0 papillary thyroid carcinoma patients who underwent unilateral prophylactic central neck dissection.Reference Lang, Wong, Wan and Lo 20 The median numbers of retrieved and metastatic lymph nodes were 8 (range, 3–26) and 4 (range, 1–16), respectively. They found that lymph node ratio was associated with higher post-ablative thyroglobulin levels, which may lead to a greater recurrence rate. However, no direct association between lymph node ratio and recurrence-free survival was found. This is the only study to investigate the prognostic value of lymph node ratio in patients with clinically staged N0 disease.
In a study, by Schneider et al., of 69 patients who underwent therapeutic neck dissection for N1 disease, 35 patients (50.7 per cent) underwent therapeutic central neck dissection for N1a disease.Reference Schneider, Mazeh, Chen and Sippel 21 Only lymph node ratio was significantly associated with recurrence on multivariate analysis, with an average hazard ratio of 19.5 (95 per cent CI = 4.1–22.9, p < 0.01) in patients with a central lymph node ratio of 0.86 or more.
Ryu et al. showed that a central lymph node ratio of 0.65 or more was the only independent variable predictive for recurrence in a multivariate analysis of 295 patients with N1a papillary thyroid carcinoma who underwent thyroidectomy and bilateral central neck dissection.Reference Ryu, Song, Choi, Roh, Nam and Kim 22 They found no association between recurrence-free survival and tumour size, multi-centricity, thyroiditis and total number of positive lymph nodes. The 10-year estimated recurrence-free survival rates were 98.6 per cent and 75.4 per cent for patients with lymph node ratios of less than 0.65 and 0.65 or more, respectively. The median numbers of retrieved and metastatic lymph nodes were 7 (range, 1–30) and 2 (range, 1–15), respectively.
Chang et al. investigated 192 patients with metastatic papillary thyroid carcinoma who underwent therapeutic central neck dissection.Reference Chang, Kim, Jung, Kim, Lee and Bae 23 The median numbers of retrieved and metastatic lymph nodes were 7 (range, 1–38) and 2 (range, 1–15), respectively. Multivariate analysis showed that a lymph node ratio of 0.48 or more was not an independent risk factor for recurrence. Instead, they found that maximal size of metastatic foci, American Thyroid Association risk categories and stimulated serum thyroglobulin levels were predictive of recurrence after surgery. A pre-ablation thyroglobulin level of 9.3 ng/ml or higher was an independent risk factor for recurrence.
Central and lateral compartment analysis
Four out of eight studiesReference Schneider, Mazeh, Chen and Sippel 21 , Reference Vas Nunes, Clark, Gao, Chua, Campbell and Niles 24 – Reference Lee, Roh, Gong, Cho, Choi and Nam 26 did not differentiate between N1a and N1b patients when investigating the effect of lymph node ratio on recurrence-free survival. Three of these studies (75 per cent) found a significant association between lymph node ratio and recurrence-free survival on multivariate analysis.
As mentioned above, Schneider et al. conducted a study that included 69 patients who underwent therapeutic neck dissection for N1 disease (34 patients underwent central and lateral neck dissection).Reference Schneider, Mazeh, Chen and Sippel 21 The median numbers of retrieved and metastatic lymph nodes were 20 (range, 3–94) and 4 (range, 0–29), respectively. A lymph node ratio of 0.7 or higher was the only significant risk factor for recurrence on multivariate analysis, with a hazard ratio of 19.5 (95 per cent CI = 4.1–22.9, p < 0.01). All other prognostic factors, including age, gender, tumour size, multifocality, extrathyroid extension, lymphovascular invasion, extranodal extension and radioactive iodine, failed to predict recurrence on multivariate analysis.
In a study, by Vas Nunes et al., of 198 patients with papillary thyroid carcinoma undergoing prophylactic and therapeutic neck dissection, 60 per cent of the patients had central compartment dissection, and 40 per cent had either lateral compartment dissection or both central and lateral compartment dissection.Reference Vas Nunes, Clark, Gao, Chua, Campbell and Niles 24 Lymph node ratio was associated with a decrease in recurrence-free survival, with a hazard ratio of 3.2 (95 per cent CI = 1.4–7.3, p = 0.005). Patients with a lymph node ratio of 0.3 or more had a 3.4 times higher risk of persistent or recurrent disease compared to patients with a lymph node ratio of 0, while patients with a lymph node ratio of 0.11 or less had an 80 per cent chance of remaining disease-free over five years.
Park et al. found that loco-regional recurrence-free survival was significantly decreased in patients with more than six metastatic lymph nodes and a lymph node ratio of more than 0.22.Reference Park, Wang, Shin, Kim, Son and Lee 25 The five-year recurrence-free survival rate for patients with a lymph node ratio of 0.22 or less or a ratio of more than 0.22 was 97.1 per cent and 78.8 per cent, respectively. Their study involved 212 papillary thyroid carcinoma patients with N1b disease. The average number of removed lymph nodes per patient was 31.4.
Lee et al. recently published a study on the prognostic value of the lymph node ratio of lateral compartment metastases in papillary thyroid carcinoma on loco-regional recurrence.Reference Lee, Roh, Gong, Cho, Choi and Nam 26 They failed to show a significant association between lymph node ratio and recurrence-free survival in 136 patients with papillary thyroid carcinoma and clinically staged N1b disease who underwent thyroidectomy with therapeutic central and lateral neck dissection. However, other prognostic factors, such as largest size of metastatic lymph node, American Thyroid Association risk categories, and stimulated thyroglobulin level, were independent predictive factors of recurrence-free survival on multivariate analysis. The median numbers of dissected and positive lymph nodes were 48 (range, 12–144) and 10 (range, 1–48), respectively.
General discussion
As papillary thyroid carcinoma is an indolent tumour with an excellent prognosis, it is difficult to adequately evaluate the prognostic significance of lymph node ratio on survival unless data from a large patient population followed for a long period are available for analysis. Only one study investigated the effect of lymph node ratio on disease-specific mortality.Reference Schneider, Chen and Sippel 18
Prophylactic central neck dissection in clinically staged N0 papillary thyroid carcinoma patients is still controversial. Lang et al. showed no association between lymph node ratio and loco-regional recurrence in 51 patients with clinically staged N0 disease.Reference Lang, Wong, Wan and Lo 20 A larger cohort is required to better understand the effect of central compartment micro-metastasis on prognosis.
While the majority of the studies reviewed showed a positive association between lymph node ratio and higher loco-regional recurrence on multivariate analysis, many of these studies failed to show a significant association between recurrence and some of the popular prognostic factors used today, such as the American Joint Committee on Cancer TNM staging system, tumour size, extrathyroid extension, age or gender. This may indicate the need for a new staging system for papillary thyroid carcinoma that incorporates regional metastatic burden in its classification system, to better evaluate recurrence risk in patients with pathologically staged N1 disease. Patients with a higher lymph node ratio should be considered ‘high risk’ for loco-regional recurrence and should be monitored closely during follow up. Early detection of recurrence may lead to effective and overall better management.
We recognise several significant drawbacks in our review. All studies were retrospective, and the majority did not differentiate between therapeutic and prophylactic neck dissections, or between central and lateral compartment analysis. All studies but one included a relatively small number of patients and short follow-up periods, limiting the studies’ ability to assess the relationship between lymph node ratio and disease-related survival. Lymph node ratio is affected by the extent of surgery (number of lymph nodes retrieved) and by the enthusiasm of the pathologist, which may greatly alter the result of lymph node ratio. Furthermore, the extent and dosage of radioactive iodine used post-operatively was not included in the multivariate analysis in the majority of the studies.
Future prospective studies examining the relationship of lymph node ratio to recurrence and survival should take into account the laterality of the lymph nodes, their size, and the distinction between micro- and macro-metastases.
Conclusion
Regional lymph node ratio in papillary thyroid carcinoma patients with pathologically staged N1 disease was an independent predictor for loco-regional recurrence in 60 per cent and 75 per cent of the studies that investigated the effect of central compartment metastases and both central and lateral compartment metastases on recurrence-free survival, respectively. Many of the well-accepted prognostic factors, such as the American Joint Committee on Cancer TNM classification, extrathyroidal extension, age and gender, failed to predict recurrence on multivariate analysis in many of the studies. A newer classification system that incorporates lymph node ratio as a prognostic factor for recurrence should be considered.
A high lymph node ratio should encourage clinicians to closely follow up metastatic papillary thyroid carcinoma patients. Based on this review, we recommend the close follow up of patients with a lymph node ratio of more than 0.3 in both central and lateral compartment metastases because of the increased risk for recurrence. Lymph node recurrence in the lateral compartment only had an effect on survival. As not all studies reached statistical significance for the association between lymph node ratio and recurrence, large randomised controlled studies should be conducted to better elucidate the prognostic value of the lymph node ratio in metastatic papillary thyroid carcinoma.
