10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Received: 1 August 2020 Revised: 30 September 2020 Accepted: 30 October 2020 DOI: 10.1002/hed.26541 CLINICAL REVIEW Prognostic role of intraparotid lymph node metastasis in primary parotid cancer: Systematic review Orlando Guntinas-Lichius MD1,2,3 | Jovanna Thielker MD1,2 | K. Thomas Robbins MD, FRCSC, FACS4 | Kerry D. Olsen MD5 | Ashok R. Shaha MD6 | Antti A. Mäkitie MD, PhD7,8,9 | Remco de Bree MD, PhD10 | Vincent Vander Poorten MD, PhD, MSc3,11 | Miquel Quer MD, PhD3,12 | Alessandra Rinaldo MD, FRCSEd ad hominem, FRCS (Eng, Ir) ad eundem, FRCSGlasg, FACS13 | Luiz Paulo Kowalski MD, PhD14 | Juan Pablo Rodrigo MD, PhD15 | Marc Hamoir MD16 | Alfio Ferlito MD, DLO, DPath, FRCSEd ad hominem, FRCS (Eng, Glasg, Ir) ad eundem, FDSRCS ad eundem, FACS, FHKCORL, FRCPath, FASCP, IFCAP17 1 Department of Otorhinolaryngology, Jena University Hospital, Friedrich Schiller University, Jena, Germany 2 Facial Nerve Center, Jena University Hospital, Friedrich Schiller University, Jena, Germany 3 Multidisciplinary Salivary Gland Society, Geneva, Switzerland 4 Department of Otolaryngology-Head and Neck Surgery, Southern Illinois University, Springfield, Illinois 5 Department of Otorhinolaryngology, Mayo Clinic, Rochester, Minnesota 6 Head and Neck Service, Memorial Sloan-Kettering Cancer Center, New York, New York 7 Department of Otorhinolaryngology-Head and Neck Surgery, University of Helsinki and Helsinki University Hospital, Helsinki, Finland 8 Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland 9 Division of Ear, Nose and Throat Diseases, Department of Clinical Sciences, Intervention and Technology, Karolinska Institute and Karolinska Hospital, Stockholm, Sweden 10 Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands 11 Otorhinolaryngology-Head and Neck Surgery, University Hospitals Leuven and Department of Oncology, Section Head and Neck Oncology, Leuven, Belgium 12 Department of Otolaryngology, Hospital Santa Creu i Sant Pau, Barcelona, Spain 13 University of Udine School of Medicine, Udine, Italy 14 Department of Head and Neck Surgery and Otorhinolaryngology, A.C. Camargo Cancer Center, and Division of Head and Neck Surgery, Sao Paulo State University Medical School, S~ao Paulo, Brazil 15 Servicio de Otorrinolaringología, Hospital Universitario Central de Asturias, Instituto Universitario de Oncología del Principado de Asturias, Universidad de Oviedo, Oviedo, Spain 16 Department of Head & Neck Surgery, St Luc University Hospital and King Albert II Cancer Institute, Brussels, Belgium; Institut de Recherche Expérimentale et Clinique (IREC), Université Catholique de Louvain, Brussels, Belgium 17 Coordinator of the International Head and Neck Scientific Group, Padua, Italy This paper was written by members and invitees of the International Head and Neck Scientific Group (www.IHNSG.com) This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2020 The Authors. Head & Neck published by Wiley Periodicals LLC. Head & Neck. 2021;43:997–1008. wileyonlinelibrary.com/journal/hed 997 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 998 GUNTINAS-LICHIUS ET AL. Correspondence Orlando Guntinas-Lichius, MD, Abstract Department of Otorhinolaryngology, Jena Background: The prognostic importance of intraparotid lymph node metasta- University Hospital, Am Klinikum 1, D- sis (P+) in patients with primary parotid gland carcinoma is unclear. 07747 Jena, Germany. Email: orlando.guntinas@med.uni-jena.de Methods: Nineteen retrospective and noncomparative cohort studies, publi- shed between 1992 and 2020, met the inclusion criteria and included 2202 Section Editor: Eisele David patients for this systematic review. Results: The pooled prevalence of the P in adult patients in the unselected studies was 24.10% (95% confidence interval = 17.95-30.25). The number of P+ lymph nodes per patient was counted in only three studies and ranged from 1 to 11. The 5-year recurrence-free survival rate based on Kaplan-Meier analy- sis varied from 83% to 88% in P− patients compared to 36% to 54% in P+ patients. The average hazard ratio for tumor recurrence in patients with P+ compared to P− was 2.67 ± 0.58. Conclusions: P+ is an independent negative prognostic factor in primary parotid gland cancer and should be included into the treatment planning. KEYWORDS clinical significance, intraparotid lymph node metastasis, nodal metastasis, prevalence, primary parotid cancer, survival 1 | INTRODUCTION node disease.14 The protocols for managing patients with positive intraparotid lymph nodes are not well defined During embryology, the lymphatic system of the neck including the issue of staging and therapy. For example, develops after encapsulation of the submandibular and should one positive intraparotid lymph node <3 cm in a sublingual glands but before encapsulation of the parotid case of a metastasis-free neck classified as N1? Should glands. This is why the submandibular and sublingual several positive intraparotid lymph nodes <6 cm in a glands do not contain intraglandular lymph nodes, metastasis-free neck be classified as N2b? Finally, there whereas the parotid glands contain lymph nodes are no standards for pathologic analysis and reporting of throughout the gland.1,2 Cadaver dissections have shown a parotid gland specimen with a primary cancer. Thera- that the number of lymph nodes in the superficial parotid peutic guidelines for P+ primary salivary gland cancer lobe ranges from 0 to 22 and in the deep lobe from are needed. 0 to 4.3-6 Some of the intraparotid lymph nodes (P) can This systematic study sought to review the current also be identified by lymphatic mapping during a parotid evidence for the role of P+ in primary parotid cancer. sentinel node procedure.7 For regional cutaneous squa- The main outcome measure was the prevalence of P+ mous cell carcinoma and melanoma, it has been well and secondary outcomes were the number of positive known for a long time that metastasis to intraparotid intraglandular lymph nodes, the relationship of P+ to lymph nodes (P+) is a negative prognostic factor com- other oncological and histopathological parameters espe- pared to the absence of intraparotid lymph nodes (P−), cially local recurrence and survival. all of which are independent of neck lymph node sta- tus.8,9 However, these lymph nodes can also be associated with primary parotid cancer. In recent years, studies of 2 | MATERIALS AND METHODS parotid cancer10-13 have reported the metastatic rate to intraparotid lymph nodes and concluded that P+ is an 2.1 | Literature search strategy independent risk factor for poor outcomes for primary parotid cancer. The current UICC staging system for local This systematic review followed The Preferred Reporting lymph node metastasis (N) does not differentiate between Items for Systematic Reviews and Meta-Analyses the presence of cervical nodal disease vs parotid lymph (PRISMA) guidelines.15 An online literature search was 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GUNTINAS-LICHIUS ET AL. 999 performed using the PubMed, SCOPUS, Embase, Web of for each study was calculated as the sum of all scores, Science, Google Scholar, and Cochrane Central Register thus ranging between 0 and 12 points where higher of Controlled Trials databases from the date of first publi- scores indicated better quality. No weighting was used. cation in these databases to June 30, 2020. The search strategy aimed to capture all papers on intraparotid lymph nodes involvement and metastases in parotid can- 2.4 | Outcome measurements and cer. The search terms were “primary parotid cancer,” statistics “primary parotid carcinoma,” “parotid neoplasm,” “intraparotid lymph node,” “parotid lymph nodes,” For the primary outcome, we calculated the prevalence “intraglandular,” “lymphatic metastasis,” and “lymph of intraparotid metastasis. Secondary outcome measures nodes.” Two authors (OGL, JT) independently reviewed were the number of intraparotid lymph node metastases; the reference lists, the titles, and abstracts for potentially the association between P+ and other tumor-related eligible studies. Both review authors independently parameters (T and N classification, perineural invasion, assessed each full-text paper for eligibility and then positive margins, lymphovascular invasion, tumor grade); selected the studies for inclusion. There were no disagree- and finally the impact of P+ on recurrence-free survival ments about inclusion. and overall survival. All descriptive statistical analyses were performed using the IBM SPSS program, version 25. Data from each study on the P+ rate were extracted and 2.2 | Selection criteria recorded in an Excel (Microsoft) spreadsheet. The pooled prevalence of P+ with 95% confidence interval (CI) was Studies were selected based on the ideas behind the PICO calculated. The results are presented with a forest plot. model.16 The C (comparison) of the PICO concept was Statistical heterogeneity was calculated with the Higgins excluded since comparison studies are not existing on the I2 statistics19 Other data (number of P+, follow-up time, topic and were therefore not the focus for this systematic survival rates, hazard ratios) are expressed as mean ± SD review. Studies had to meet the following inclusion and ranges. criteria: (a) English language; (b) primary presentation of cancer to the parotid gland; (c) primary parotid cancer; (d) no prior treatment of the parotid gland or neck for 3 | RESULTS malignancy; (e) detailed surgical description of the affected parotid gland and its outcomes; and (f) mean The search identified 139 articles. Duplicates were follow-up of >24 months. Exclusion criteria included removed. From the remaining 100 abstracts, 80 publica- (a) studies with non-primary tumors; (b) no surgical tions were excluded Table S1). The full texts of the treatment of the affected parotid gland; (c) prior treat- remaining 20 articles were reviewed. Nineteen studies ment of the affected parotid gland or prior treatment of were included (Figure 1).11-13,20-35 There were no the neck; (d) cohort studies with less than 20 patients; and (e) case reports. 2.3 | Quality assessment Two authors (OGL, JT) independently evaluated the methodological quality of the included studies. For evalu- ation, the criteria from the “guidelines for assessing the quality in prognostic studies on the basis of Framework of Potential Biases” was used.17,18 These quality criteria included six domains of potential bias with prognostic studies: study population, study attrition, measurement of prognostic factors, measurement of outcomes, mea- surement of and controlling for confounding variables, and analysis approaches. The reviewers graded each cri- terion as yes = 2, partly = 1, or no/unclear = 0. In case of disagreement between the two reviewers, the study was F I G U R E 1 Literature search flow chart according to the discussed until a consensus was found. A quality score PRISMA guidelines15 TABLE 1 Characteristics and findings of the included studies 1000 Subtypes Follow up of parotid time Outcome RFS Outcome Outcome OS Study Year n Population cancer (months) P+% P+ (n) P+ association rate % OS rate % Outcome RFS P+ P+ Armstrong 1992 30 1939-1982 All, 90 with ND; — 53.3 in cN0 — — — — — — et al 20 focus on cN0 Cheng et al 2019 249 2002-2015 All, focus on Mean 89.4 18.1 — — — — — HR 1.845 21 neutrophil-to- (range, 8- (1.054- lymphocyte ratio 190) 4.531) P = .003 Fang et al 2019 53 2005-2016 SCC Mean 67.3 43.3 — — — — HR 2.347 Not significant 22 (range, 4- (CI 1.279-5.612) 135) P = .007 Fang et al 2019 144 2000-2014 AcCC focus on rM Mean 90.7 31.8 — — — — — Distant 23 + (range, 16- metastasis 211) survival P+ (1.854 [1.061- 4.144], P = 0.011 Fang et al 2019 73 1990-2017 Pediatric MEC Mean 95.4 17.8 HR 2.223 HR 1.897 24 (range, 13- (CI 1.658-5.897) [1.142- 286) P < .001 4.656]) P < .001 Feng et al 2019 337 2000-2017 All Mean 71.7 32.9 Mean 1.4 Higher T classification, higher All 96 5y — P+ ≤ 2 LN — 13 (range, 4- (range, 1-5). pN classification, perineural all 94 10y RR 6.175 209) invasion, R+, P+ ≤2 LN CI 2.797-13.634) lymphovascular invasion 65 5y P < .001 56 10y P+ > 2 LN P+ >2 LN RR 21.563 (CI 7.450- 22 5y 62.415) 22 10y P < .001 Grasl et al 2019 66 1998-2017; AcCC Mean 55.5 7.6 — No influence on recurrence, — — — — 25 1989- (median no influence on grading 2005 46.7) Klussmann 2008 142 1986-2006 All Median 24.4 17.6 Range 0-11 — P+ not significant P+ not — — et al 11 (range, 0.2- significant 250.5). Lim et al 12 2014 86 1998-2009 All Median 42 38.7 N classification — P− 75 3y — — cN0 P+ 34.2 3y median 36 P = 0.0037 cN+ median 38 P− median 34 P+ GUNTINAS-LICHIUS ET AL. 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License TABLE 1 (Continued) Subtypes Follow up of parotid time Outcome RFS Outcome Outcome OS Study Year n Population cancer (months) P+% P+ (n) P+ association rate % OS rate % Outcome RFS P+ P+ Lima et al 2005 126 1974-1998 All Mean 60 12.0 — — — — — — 26 (range, 4- GUNTINAS-LICHIUS ET AL. 216) Nisa et al 27 2015 95 1997-2010 All Median 25.26 Median 1; Higher pT, higher pN, higher P+ 36 5y P+ 70 5y — — 38.41 mean 2.91; overall stage; high-risk P- 85 5y P- 78 5y (range, range, 1–10). carcinomas, treatment 0.46- failures 160.75) Niu et al 28 2019 190 2001-2017 MEC Mean 71.1 24.7 — Higher T classification, higher P- 88 10y — HR 2.357 — N classification, perineural P+ 54 10y (CI 1.163-4.776) invasion, high grade, P < .001. P = .017 lymphovascular invasion Paderno 2018 104 1997-2017 All Median 51 20.2 — Perineural invasion, higher 66.9 5y 74.7 5y RR 3.64; P = .00102 RR 2.97; et al29 grade, lateral lymph node P = .00881 involvement, extranodal extension Schneider 2020 50 1993-2010 All Mean 98.2 38.8 — — P+ 50 5y P+ not — — et al 30 (range, 4- P− 80 5y significant 275) Seng et al 2020 77 2000–2017 Pediatric Mean 85.0 19.5 — Higher tumor stage, high- All 91 10y — HR 2.805 — 31 (range, 13- grade tumor, P+ 36 5y (CI 1.697-5.119) 173) P− 97 5y P < .004 Shang et al 2019 122 2000-2016 High-grade MEC Mean 69.7 48.4 — Higher T classification, higher P− 83 5y P+ superficial lobe 32 (range, 5- N classification P+ only superficial parotid HR 2.142 (CI 201) lobe parotid 56 0.952-4.819) 5y P = .065 P+ only deep P+ deep lobe parotid lobe parotid HR 5.659 metastasis, 24 5y (CI 1.958-16.355) P+ superficial and P = .001 deep lymph P+ superficial node parotid and deep parotid lobe 11 5y HR 13.517 P < .001 (CI 4.934-37.033) P < .001 Stenner 2012 70 1987-2009 All, T1-2 Median 51.7 73.3 — — — — — — et al 33 (range, 0.2- 191.5) Stodulski 2017 66 1992-2010 All, focus on cN0 Mean 94.8 13.6 — Occult N+ — — — — et al 34 (range, 60- 288) 1001 (Continues) 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1002 GUNTINAS-LICHIUS ET AL. prospective studies identified. Instead, all studies were Abbreviations: 5y, 5 year; 10y, 10 year; AcCC, acinic cell carcinoma; CI = 95% confidence interval; HR, hazard ratio; LN, lymph node; MEC, mucoepidermoid carcinoma; ND, neck dissection; OS, overall survival; P+, intraparotid Outcome OS retrospective and noncomparative trials. The 19 studies included a total of 2202 patients (Table 1). P+ — 3.1 | Quality scores Outcome RFS P+ (CI 0.913-2.116) (CI 1.039-3.668) (CI 1.229-7.194) For the 19 selected studies, the quality scores ranged P = 0.011 HR 1.338 HR 1.992 HR 3.118 P = .063 P = .001 from 6 to 9 points (Table S2). Twelve studies were classi- fied medium quality (5-8 points) and seven studies as P1 P2 P3 high quality (9-12 points). It is important to note that nine of the studies came from the same institution, the OS rate % Outcome Zhengzhou University, China, with varying time periods, — inclusion criteria, and the outcome mea- 13,21-24,28,31,32,35 sures. All but one study were single insti- tutional noncomparative cohort studies. One study Outcome RFS collected data from two centers.25 Twelve studies rate % included all histological types of primary parotid cancer,11-13,20,21,26,27,29-31,33,34 four studies were focused — on mucoepidermoid cancer,24,28,32,35 two studies on lymph node metastasis; P−, no intraparotid lymph node metastasis; RFS, recurrence-free survival; RR, relative risk; SCC, squamous cell carcinoma. Higher T classification, higher N classification, perineural acinic cell carcinoma, 23,25 and one study on squamous lymphovascular invasion invasion, high grade, cell carcinoma.22 Two studies reported only pediatric cases24,31 and one study focused only on early-stage (low- P+ association volume) cancer (T1/T2 N0M0).33 Because parotid cancer is relatively rare, long inclusion periods were chosen in all studies. Important limitations were the variability of the extent of the parotidectomy, unclear evaluation of the intraparotid lymph nodes, limited, or inconsistent follow- up periods, and small subgroups limiting the explanatory P+ (n) power and the multivariate analyses. — 3.2 | Primary outcome: Prevalence of P+ P+% 24.6 The mean relative number of P+, that is, the relation of Follow up (months) (range, 6- positive nodes to all evaluated intraparotid lymph nodes Mean 77 147) time was 29.55 ± 16.53% (range, 7.6%-73.3%). The pooled prev- alence of the P+ rate in adult patients in the unselected studies was 24.10% (CI = 17.95-30.25; Figure 2). The pooled prevalence of the P+ rate in adult patients in dif- of parotid Subtypes ferent subtypes of parotid cancer was more variable with cancer MEC 29.00% (CI = 17.59-40.42; Figure 3). Only three studies stated the absolute number of nodes for each P+ Population case. 11,13,28 Herein, the number of P+ in each cohort 2005-2018 ranged from 1 to 11. P+ was detected in both the superfi- (Continued) cial lobe and the deep lobe. 122 n Year 2020 3.3 | Association of P+ with other tumor characteristics TABLE 1 Wu et al35 Study Associations with other tumor characteristics were ana- lyzed with univariate analyses in 10 studies (cf. Table 1). 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GUNTINAS-LICHIUS ET AL. 1003 F I G U R E 2 Forest plot of pooled prevalence of P+ in unselected studies on adult patients with primary parotid cancer. Weights are from random effect analysis. C, confidence interval [Color figure can be viewed at wileyonlinelibrary.com] In selected studies, P+ was more common with a higher patients with P+ compared to P− was 2.67 ± 0.58 (range, T classification, higher pN classification, finding of peri- 2.22-3.64). Only one study presented 5-year overall sur- neural invasion, positive margins, lymphovascular inva- vival rates. Here, patients with P+ had a 5-year survival sion, and high-grade tumors. P+ had a higher risk of rate of 70% compared to 78% in P− patients. The average occult lymph node metastasis in the neck (N+) in one calculated hazard ratio for a risk of death in patients with study.34 P+ compared to P− was increased by 2.14 ± 0.55 (range, 1.85-2.97) in four studies.21,23,24,29 However, one study did not show a significant difference in the risk for death 3.4 | Impact of P+ on recurrence-free between P+ and P− patients in the Cox regression survival and overall survival analysis.22 The follow-up time of individual patients varied from 0.2 to 288 months. Only six studies reported median follow- 4 | DISCUSSION up time data.11,12,25,27,29,33 The average reported that median follow-up was 39.1 ± 9.9 months (range, The parotid gland has between 0 and 26 intraparotid 24.4-51.7). The remaining 13 studies provided the mean lymph nodes.3,6 Many patients did not have P+ metasta- values of follow-up time: 78.9 ± 14.2 months (range, sis. Hence, intraparotid lymph node metastasis does not 55.5-98.2). Nine studies reported data on recurrence-free seem to be first echelon lymph nodes for regional spread- survival.13,22,24,27-29,31,32,35 Nine studies provided data on ing in all cases. It may be that the location of the primary overall survival.11,12,21-24,27,29,30 P+ was a significant neg- tumor in the parotid gland influences if an intraparotid ative prognostic factor for survival in all the multivariate lymph node metastasis can take place or not because the regression models except one. The 5-year recurrence-free lymphatics flow from superior to inferior portion of survival rate in Kaplan-Meier analysis varied from 83% to the gland. It is imaginable that if the tumor is located in 88% (97% in children) in P− patients compared to 36% to the far inferior part of parotid gland, level IIa or IIb 54% in P+ patients. If >2 lymph nodes were positive and lymph node metastasis might regularly occur without if lymph nodes in the deep lobe were involved, the 5-year intraparotid lymph node metastasis. The presented data recurrence-free survival rate dropped to 11% to 22%. The are too sparse allowing a detailed analysis of the average calculated hazard ratio for tumor recurrence in intraparotid lymphatic metastasis routes. Furthermore, 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1004 GUNTINAS-LICHIUS ET AL. F I G U R E 3 Forest plot of pooled prevalence of P+ in studies in adult patients focused on a histological subtype of primary parotid cancer. Weights are from random effect analysis. AcCC, acinic cell carcinoma; CI, confidence interval; MEC, mucoepidermoid carcinoma; SCC, squamous cell carcinoma [Color figure can be viewed at wileyonlinelibrary.com] only a few of the presented studies undertook the effort recurrence than patient with P−. The relationship of P+ to distinguish between metastasis in the superficial and to overall survival was analyzed in only four stud- the deep lobe of the parotid gland. It can only be stated ies.21,23,24,29 In these reports, P+ had a 2-fold higher risk that in some patients with P+ in the superficial gland, of death compared to P−. Overall, the data on the 2202 regional metastasis to the upper neck cervical lymph patients included into this review suggest that P+ is an nodes occurred without P+ in the deep lobe, that is, important independent predictor of a worse outcome in bypassing the deep parotid gland. There is some limited patients with primary parotid cancer. evidence that P+ in the deep lobe has a worse prognostic The significance of intraparotid lymph node metasta- than P+ in the superficial lobe. sis has previously been examined for cutaneous squa- Prospective studies using serial section histology to mous cell cancer (CSCC). O'Brien et al showed that P+ is detect all possible intraparotid lymph nodes and the true an independent predictor for a worse outcome in CSCC, finding of P+ in primary parotid cancer are lacking. Of independent of the neck status.8 This study led to the the retrospective studies analyzed for the present system- proposal of a revised staging system dividing the regional atic review, only three of the studies provided an absolute lymph node staging system into parotid and cervical dis- number positive lymph nodes among the patients with ease, defined as “P” and “N” classifications (Table 2). P+ disease.11,13,28 All included studies reported the pro- Also the group at the Mayo clinics reported on the impor- portion of positive lymph nodes among P+ patients, tance of this concept.9 The attempt to demonstrate the which ranged from 7.6% to 73.3% with a pooled preva- independent outcome influence of intraparotid vs neck lence for the unselected studies in adult patients of lymph nodes has so far not been incorporated into the 24.10% (CI = 17.95-30.25). Taking into account the meth- AJCC nodal staging of metastatic CSCC.36,37 Also for pri- odological limitations of the included studies, it can be mary parotid cancer, the argument has been made to postulated that up to a third of all intraparotid lymph consider the intraparotid lymph nodes as an important nodes in patients with primary parotid cancer may be entity.10,38 The results of the present meta-analysis sug- positive for metastatic cancer. Most studies showed that gest denoting the intraparotid and neck lymph node the presence of P+ was an independent predictor for metastases separately for improved staging of primary lower recurrence-free survival.13,22,24,28,29,31,32,35 Patients parotid cancer (Table 2). Wu et al classified 122 patients with P+ had a 2- to 3-fold higher risk for tumor with parotid mucoepidermoid cancer (P0: n = 92, P1: 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GUNTINAS-LICHIUS ET AL. 1005 n = 10, P2: n = 16; P3: n = 4; definition of P classifica- Future studies should focus on histopathology stan- tion, see Table 2).35 Although the sample size was lim- dards defining the assessment of intraglandular lymph ited, they showed that the P classification of the nodes in histopathology reports. The parotid surgeon intraparotid nodes was an independent prognostic factor should ask their pathologist to routinely look for for recurrence, independent from the N classification. intraparotid lymph nodes in all the parotid specimen.39 This systematic review has several limitations. All Consensus papers for histopathology reporting of parotid included trials were retrospective noncomparative cohort gland neoplasms typically point out the importance of studies. The data did not allow a formal meta-analysis. the lymph node metastasis but do not give any guidance Therefore, selection bias cannot be ruled out. The extent on how to classify positive intraparotid lymph nodes in of the parotidectomy procedures and the histopathologi- the current TNM staging system.40 Most guidelines rec- cal analysis of the intraparotid lymph nodes were not ommend at least a total parotidectomy at least for high- standardized or the details were not presented in these grade parotid cancer.41,42 Total parotidectomy is the only studies. Nine of the 19 included studies came from the way to assess the deep parotid nodes.42 same institution in China,13,21-24,28,31,32,35 and several of In a recent systematic review based on nine retrospec- these studies focused more narrowly on specific histo- tive studies (as no prospective trials were available), the logic cancers.22-25,28,32,35 This makes it likely that there question of performing a partial parotidectomy for a pre- was an overlap of patients among these nine studies. sumably benign tumor that proves to be a T1 low-grade T A B L E 2 Proposal for an improved staging of nodal metastasis of primary parotid cancer based on the American Joint Committee on Cancer,14 the O'Brien revised staging system,8 and the proposal made for metastatic cutaneous squamous cell cancer37 AJCC 8th edition Proposed revised staging system N classification P classification N classification NX = regional lymph nodes cannot PX = parotid lymph nodes cannot be NX = regional lymph nodes cannot be be assessed assessed assessed N0 = no regional lymph node P0 = no intraparotid metastases N0 = no regional lymph node metastases metastases N1 = metastasis in a single ipsilateral P1 = metastatic node up to 3 cm in N1 = metastasis in a single ipsilateral lymph node, 3 cm or smaller in diameter in the parotid gland lymph node, 3 cm or smaller in greatest dimension and ENE(−) greatest dimension and ENE(−) N2a = metastasis in a single P2 = metastatic node more than 3 cm N2a = metastasis in a single ipsilateral ipsilateral lymph node larger than up to 6 cm diameter or multiple lymph node larger than 3 cm but not 3 cm but not larger than 6 cm in parotid nodes larger than 6 cm in greatest greatest dimension and ENE(−) dimension and ENE(−) N2b = metastasis in multiple P3 = metastatic node more than 6 cm N2b = metastasis in multiple ipsilateral ipsilateral nodes, none larger than in diameter or disease involving the nodes, none larger than 6 cm in 6 cm in greatest dimension and facial nerve or skull base greatest dimension and ENE(−) ENE(−) N2c = metastasis in bilateral or N2c = metastasis in bilateral or contralateral lymph nodes, none contralateral lymph nodes, none larger than 6 cm in greatest larger than 6 cm in greatest dimension and ENE(−) dimension and ENE(−) N3a = metastasis in a lymph node N3a = metastasis in a lymph node larger than 6 cm in greatest larger than 6 cm in greatest dimension and ENE(−) dimension and ENE(−) N3b = metastasis in a lymph node N3b = metastasis in a lymph node more more than 3 cm in greatest than 3 cm in greatest dimension with dimension with extranodal extranodal extension (ENE+) or extension (ENE+) or multiple multiple ipsilateral, or any ipsilateral, or any contralateral, or contralateral, or bilateral node(s) with bilateral node(s) with extranodal extranodal extension (ENE+) extension (ENE+) Abbreviations: AJCC, American Joint Committee on Cancer; ENE, extranodal extension. 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1006 GUNTINAS-LICHIUS ET AL. malignancy with free margins in the final histopathology intraparotid lymph node metastases in its associated clas- was raised based on the indication for removing all sification of lymph node disease. lymph nodes at risk.43 Following the results of the pre- sent study, such limited approaches for parotid cancer ACKNOWLEDGMENT have to be applied with caution. Much confusion would Open access funding enabled and organized by be eliminated, and guidelines as to when to remove all Projekt DEAL. parotid tissue would be clearer, if all parotid specimens had a complete evaluation of their nodal status. Any find- DATA AVAILABILITY STATEMENT ing of a positive node could be an indication for complet- The data that support the findings of this study are avail- ing a total parotidectomy and neck dissection. able on request from the corresponding author. All rele- The studies analyzed for the present systematic vant data is presented in the manuscript. review do not allow to draw the clear conclusion that smaller tumors have a lower risk for P+. If limited sur- gery less than a total parotidectomy is performed, an ORCID inclusion of the tumor bed and the nonresected parotid Orlando Guntinas-Lichius https://orcid.org/0000-0001- parts into the postoperative radiotherapy concept has to 9671-0784 be discussed if postoperative radiotherapy is indicated for K. Thomas Robbins https://orcid.org/0000-0002-3255- other reasons, or at least a close follow-up of the patient 8960 is recommended. In such a situation, radiotherapy might Ashok R. Shaha https://orcid.org/0000-0001-7478-9436 Remco de Bree https://orcid.org/0000-0001-7128-5814 also be an alternative to avoid a second surgery with increased risk of facial nerve damage. In the absence of Vincent Vander Poorten https://orcid.org/0000-0003- prospective studies, the question of whether radiotherapy 1341-829X Luiz Paulo Kowalski https://orcid.org/0000-0002-0481- or surgical therapy of intraparotid lymph node metasta- ses has therapeutic efficacy for the patient remains 156X completely unclear.44 However, considering the higher Juan Pablo Rodrigo https://orcid.org/0000-0003-3063- risk of recurrence and poor salvage of recurrence with 0890 positive intraparotid lymph nodes, strong consideration Marc Hamoir https://orcid.org/0000-0001-8547-2768 should be given for postoperative radiation therapy. Alfio Ferlito https://orcid.org/0000-0002-8247-8002 Whether the entire neck should be included in postopera- RE FER EN CES tive radiation therapy if the neck does not show any obvi- ous disease either on imaging or after a neck 1. Holsinger FC, Bui DT. Anatomy, function, and evaluation of the salivary glands. In: Myers EN, Ferris RL, eds. Salivary dissection remains unclear. This dilemma mirrors the Gland Disorders. Heidelberg: Springer; 2007. controversy of elective neck dissection (END) for cN0 2. Som PM, Miletich I. The embryology of the salivary glands: an parotid carcinoma, where several papers claim that elec- update. Neurographics. 2015;5(4):167-177. tive neck irradiation is as effective as END.45 3. McKean ME, Lee K, McGregor IA. The distribution of lymph nodes in and around the parotid gland: an anatomical study. Br J Plast Surg. 1985;38(1):1-5. 5 | C ON C L U S I ON 4. Garatea-Crelgo J, Gay-Escoda C, Bermejo B, Buenechea- Imaz R. Morphological study of the parotid lymph nodes. J Craniomaxillofac Surg. 1993;21(5):207-209. The average finding of P+ ranged from 7.6% to 73.3% in 5. Pisani P, Ramponi A, Pia F. The deep parotid lymph nodes: an ana- this systematic review of parotid cancers, supporting the tomical and oncological study. J Laryngol Otol. 1996;110(2):148-150. importance and need to evaluate and effectively treat the 6. Sonmez Ergun S, Gayretli O, Buyukpinarbasili N, et al. Deter- intraparotid lymph nodes. This implies a more rigorous mining the number of intraparotid lymph nodes: postmortem pathology examination with a meticulous search for examination. J Craniomaxillofac Surg. 2014;42(5):657-660. intraparotid lymph nodes in parotid cancer specimens. 7. Ollila DW, Foshag LJ, Essner R, Stern SL, Morton DL. Parotid Prospective studies are needed to further define patients region lymphatic mapping and sentinel lymphadenectomy for cutaneous melanoma. Ann Surg Oncol. 1999;6(2):150-154. at risk for P+. Such patients may benefit from a total 8. O'Brien CJ, McNeil EB, McMahon JD, Pathak I, Lauer CS, parotidectomy and should not undergo only a more lim- Jackson MA. Significance of clinical stage, extent of sur- ited parotid approach. Further study in this area should gery, and pathologic findings in metastatic cutaneous squa- also focus on the value of (adjuvant) radiotherapy for mous carcinoma of the parotid gland. Head Neck. 2002;24 positive intraparotid lymph nodes. Subsequent updates of (5):417-422. the TNM staging for salivary gland cancer should con- 9. Thom JJ, Moore EJ, Price DL, Kasperbauer JL, Starkman SJ, sider the possible benefit of including the presence of Olsen KD. The role of total parotidectomy for metastatic 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GUNTINAS-LICHIUS ET AL. 1007 cutaneous squamous cell carcinoma and malignant melanoma. 27. Nisa L, Salmina C, Dettmer MS, et al. Implications of JAMA Otolaryngol Head Neck Surg. 2014;140(6):548-554. intraglandular lymph node metastases in primary carcinomas 10. Boahene DK, Olsen KD, Lewis JE, Pinheiro AD, Pankratz VS, of the parotid gland. Laryngoscope. 2015;125(9):2099-2106. Bagniewski SM. Mucoepidermoid carcinoma of the parotid 28. Niu X, Fang Q, Liu F. Role of intraparotid node metastasis in gland: the Mayo clinic experience. Arch Otolaryngol Head Neck mucoepidermoid carcinoma of the parotid gland. BMC Cancer. Surg. 2004;130(7):849-856. 2019;19(1):417. 11. Klussmann JP, Ponert T, Mueller RP, Dienes HP, Guntinas- 29. Paderno A, Tomasoni M, Mattavelli D, Battocchio S, Lichius O. Patterns of lymph node spread and its influence on Lombardi D, Nicolai P. Primary parotid carcinoma: analysis of outcome in resectable parotid cancer. Eur J Surg Oncol. 2008;34 risk factors and validation of a prognostic index. Eur Arch (8):932-937. Otorhinolaryngol. 2018;275(11):2829-2841. 12. Lim CM, Gilbert MR, Johnson JT, Kim S. Clinical significance 30. Schneider S, Kotowski U, Kadletz L, et al. Intraparotid and cer- of intraparotid lymph node metastasis in primary parotid can- vical lymph nodes metastasis in primary parotid gland cer. Head Neck. 2014;36(11):1634-1637. carcinoma-impact on clinical outcome. Oral Surg Oral Med 13. Feng Y, Liu F, Cheng G, Fang Q, Niu X, He W. Significance of Oral Pathol Oral Radiol. 2020;129(6):570-574. intraparotid node metastasis in predicting local control in pri- 31. Seng D, Fang Q, Liu F, Luo R, Liu S. Intraparotid lymph node mary parotid cancer. Laryngoscope. 2019;129(10):2309-2312. metastasis decreases survival in pediatric patients with 14. Amin MB, Edge S, Greene F, et al. AJCC Cancer Staging Man- parotid cancer. J Oral Maxillofac Surg. 2020;78(5):852 e851- ual. 8th ed. Heidelberg: Springer; 2017. 852 e856. 15. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred 32. Shang X, Fang Q, Liu F, Wu J, Luo R, Qi J. Deep parotid lymph reporting items for systematic reviews and meta-analyses: the node metastasis is associated with recurrence in high-grade PRISMA statement. Ann Intern Med. 2009;151(4):264-269. mucoepidermoid carcinoma of the parotid gland. J Oral Maxil- W264. lofac Surg. 2019;77(7):1505-1509. 16. Cooke A, Smith D, Booth A. Beyond PICO: the SPIDER tool 33. Stenner M, Molls C, Luers JC, Beutner D, Klussmann JP, for qualitative evidence synthesis. Qual Health Res. 2012;22 Huettenbrink KB. Occurrence of lymph node metastasis in (10):1435-1443. early-stage parotid gland cancer. Eur Arch Otorhinolaryngol. 17. Hayden JA, Cote P, Bombardier C. Evaluation of the quality of 2012;269(2):643-648. prognosis studies in systematic reviews. Ann Intern Med. 2006; 34. Stodulski D, Mikaszewski B, Majewska H, Wisniewski P, 144(6):427-437. Stankiewicz C. Probability and pattern of occult cervical lymph 18. Rashid M, Kristofferzon ML, Nilsson A, Heiden M. Factors node metastases in primary parotid carcinoma. Eur Arch associated with return to work among people on work absence Otorhinolaryngol. 2017;274(3):1659-1664. due to long-term neck or back pain: a narrative systematic 35. Wu J, Fang Q, Liu F, Zhang X. Intraparotid node metastases in review. BMJ Open. 2017;7(6):e014939. adults with parotid mucoepidermoid cancer: an indicator of 19. Neyeloff JL, Fuchs SC, Moreira LB. Meta-analyses and Forest prognosis? Br J Oral Maxillofac Surg. 2020;58(5):525-529. plots using a microsoft excel spreadsheet: step-by-step guide 36. Ch'ng S, Maitra A, Allison RS, et al. Parotid and cervical nodal focusing on descriptive data analysis. BMC Res Notes. 2012; status predict prognosis for patients with head and neck meta- 5:52. static cutaneous squamous cell carcinoma. J Surg Oncol. 2008; 20. Armstrong JG, Harrison LB, Thaler HT, et al. The indications 98(2):101-105. for elective treatment of the neck in cancer of the major sali- 37. Rotman A, Kerr SJ, Giddings CEB. Elective neck dissection in vary glands. Cancer. 1992;69(3):615-619. metastatic cutaneous squamous cell carcinoma to the parotid 21. Cheng G, Liu F, Niu X, Fang Q. Role of the pretreatment gland: a systematic review and meta-analysis. Head Neck. 2019; neutrophil-tolymphocyte ratio in the survival of primary 41(4):1131-1139. parotid cancer patients. Cancer Manag Res. 2019;11:2281-2286. 38. Olsen KD, Moore EJ. Deep lobe parotidectomy: clinical ratio- 22. Fang Q, Wu J, Liu F. Oncologic outcome and potential prog- nale in the management of primary and metastatic cancer. Eur nostic factors in primary squamous cell carcinoma of the Arch Otorhinolaryngol. 2014;271(5):1181-1185. parotid gland. BMC Cancer. 2019;19(1):752. https://doi.org/10. 39. Thielker J, Grosheva M, Ihrler S, Wittig A, Guntinas-Lichius O. 1186/s12885-019-5969-6. Contemporary management of benign and malignant parotid 23. Fang Q, Wu J, Du W, Zhang X. Predictors of distant metastasis tumors. Front Surg. 2018;5:39. in parotid acinic cell carcinoma. BMC Cancer. 2019;19(1):475. 40. Seethala RR, Altemani A, Ferris RL, et al. Data set for the https://doi.org/10.1186/s12885-019-5711-4. reporting of carcinomas of the major salivary glands: explana- 24. Fang Q, Liu F, Seng D. Oncologic outcome of parotid tions and recommendations of the guidelines from the interna- mucoepidermoid carcinoma in pediatric patients. Cancer tional collaboration on cancer reporting. Arch Pathol Lab Med. Manag Res. 2019;11:1081-1085. 2019;143(5):578-586. 25. Grasl S, Janik S, Grasl MC, et al. Nodal metastases in acinic cell 41. Sood S, McGurk M, Vaz F. Management of salivary gland carcinoma of the parotid gland. J Clin Med. 2019;8(9):1315. tumours: United Kingdom National Multidisciplinary Guide- https://doi.org/10.3390/jcm8091315. lines. J Laryngol Otol. 2016;130(S2):S142-S149. 26. Lima RA, Tavares MR, Dias FL, et al. Clinical prognostic fac- 42. Olsen KD, Quer M, de Bree R, Vander Poorten V, Rinaldo A, tors in malignant parotid gland tumors. Otolaryngol Head Neck Ferlito A. Deep lobe parotidectomy-why, when, and how? Eur Surg. 2005;133(5):702-708. Arch Otorhinolaryngol. 2017;274(12):4073-4078. 10970347, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.26541 by Readcube (Labtiva Inc.), Wiley Online Library on [15/03/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 1008 GUNTINAS-LICHIUS ET AL. 43. Quer M, Olsen KD, Silver CE, et al. Is there a role for limited SU PP O R TI N G I N F O RMA TI O N parotid resections for primary malignant parotid tumors? Sur- Additional supporting information may be found online geries. 2020;1:2-9. in the Supporting Information section at the end of this 44. Vander Poorten V, Bradley PJ, Takes RP, Rinaldo A, article. Woolgar JA, Ferlito A. Diagnosis and management of parotid carcinoma with a special focus on recent advances in molecu- lar biology. Head Neck. 2012;34(3):429-440. How to cite this article: Guntinas-Lichius O, 45. Valstar MH, van den Brekel MW, Smeele LE. Interpretation of Thielker J, Robbins KT, et al. Prognostic role of treatment outcome in the clinically node-negative neck in pri- intraparotid lymph node metastasis in primary mary parotid carcinoma: a systematic review of the literature. Head Neck. 2010;32(10):1402-1411. parotid cancer: Systematic review. Head & Neck. 2021;43:997–1008. https://doi.org/10.1002/hed. 26541