European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 https://doi.org/10.1007/s00405-020-06406-9 REVIEW ARTICLE European Laryngological Society position paper on laryngeal dysplasia Part II: diagnosis, treatment, and follow‑up Hans Edmund Eckel1 · Ricard Simo2 · Miquel Quer3 · Edward Odell4 · Vinidh Paleri5 · Jens Peter Klussmann6 · Marc Remacle7 · Elisabeth Sjögren8 · Cesare Piazza9 Received: 16 March 2020 / Accepted: 16 May 2020 / Published online: 14 October 2020 © The Author(s) 2020 Abstract Purpose of review To give an overview of the current knowledge regarding the diagnosis, treatment, and follow-up of laryn- geal dysplasia (LD) and to highlight the contributions of recent literature. Summary The diagnosis of LD largely relies on endoscopic procedures and on histopathology. Diagnostic efficiency of endoscopy may be improved using videolaryngostroboscopy (VLS) and bioendoscopic tools such as Narrow Band Imaging (NBI) or Storz Professional Image Enhancement System (SPIES). Current histological classifications are not powerful enough to clearly predict the risk to carcinoma evolution and technical issues such as sampling error, variation in epithelial thickness and inflammation hamper pathological examination. Almost all dysplasia grading systems are effective in different ways. The 2017 World Health Organization (WHO) system should prove to be an improvement as it is slightly more reproducible and easier for the non-specialist pathologist to apply. To optimize treatment decisions, surgeons should know how their pathologist grades samples and preferably audit their transformation rates locally. Whether carcinoma in situ should be used as part of such classification remains contentious and pathologists should agree with their clinicians whether they find this additional grade useful in treatment decisions. Recently, different studies have defined the possible utility of different biomarkers in risk classification. The main treatment modality for LD is represented by transoral laser microsurgery. Radiotherapy may be indicated in specific circumstances such as multiple recurrence or wide-field lesions. Medical treatment currently does not have a significant role in the management of LD. Follow-up for patients treated with LD is a fundamental part of their care and investigations may be supported by the same techniques used during diagnosis (VLS and NBI/SPIES). Keywords Carcinoma in situ · Dysplasia · Laryngeal intraepithelial neoplasia · Laryngeal carcinoma · Laser surgery · Radiotherapy 6 * Elisabeth Sjögren Department of Otorhinolaryngology, Head and Neck evsjogren@lumc.nl Surgery, Medical Faculty, University of Cologne, Cologne, Germany 1 Department of Oto‑Rhino‑Laryngology, Klagenfurt General 7 Department of Otorhinolaryngology, Head and Neck Hospital, Klagenfurt am Wörthersee, Austria Surgery, CH Luxembourg, Luxembourg, Belgium 2 Department of Otorhinolaryngology Head and Neck Surgery, 8 Department of Otorhinolaryngology, Head and Neck Surgery, Guy’s and St Thomas’ Hospital, London, UK Leiden University Medical Center, Leiden, The Netherlands 3 Department of Otorhinolaryngology and Head and Neck 9 Department of Otorhinolaryngology‑Head and Neck Surgery, Surgery, Hospital de la Santa Creu i Sant Pau. Universitat ASST Spedali Civili of Brescia, University of Brescia, Autònoma de Barcelona, Barcelona, Spain Brescia, Italy 4 Department of Head and Neck Pathology, King’s College London Guy’s Hospital, London, UK 5 Department of Otorhinolaryngology Head and Neck Surgery, Royal Marsden Hospital, London, UK 13 Vol.:(0123456789) 1724 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 Diagnostic procedures have the aspect, defined by Arens and coworkers, of “per- pendicular vessels”, indicative of a potentially malignant The diagnosis of laryngeal dysplasia (LD) largely relies disease (from mild dysplasia upward) [6]. By contrast, on endoscopic procedures and on histopathology. Imaging when no vessels are visible or present a normal course modalities (CT or MR scanning) may be obtained in cases (defined by Arens as “longitudinal”, i.e., parallel to the with clinically unclear presentation, but have no role in the surface of the epithelial layer), this is strongly suggestive routine diagnostic process of suspected LD. for a benign condition or a keratotic plaque without any malignant potential [6]. Clearly, these bioendoscopic signs may now help in choosing between a watchful waiting with regular follow-up versus immediate biopsy and/or resec- Endoscopy tion, usually under general anaesthesia. Various attempts have been made to verify whether video- laryngostroboscopy (VLS) may be accurate enough in dis- Histopathological examination tinguishing intraepithelial lesions from microinvasive or frankly invasive vocal fold lesions. At least from the theo- Even leaving aside the fact that predicting the future is retical point of view, this should be possible due to the impossible, a number of practical problems make assess- well-known persistence of an undisturbed mucosal wave in ment of dysplasia and carcinoma risk in the larynx difficult. presence of mild to severe dysplasia, with impairment of There is always a risk of sampling error, because the most this physiological phenomenon when the neoplastic prolif- marked histological changes do not necessarily occur in the eration involves the vocal ligament, thus tethering the epi- red or thickly keratinised areas. Sampling to adequate depth thelium to the superficial layer of the lamina propria dur- is a particular problem when keratin is thick and the basal ing phonation. However, a number of studies have found cell layers of the epithelium may be 2 mm or more below a suboptimal accuracy of VLS in performance of such the surface. There are confusing variations in the normal a task, either due to concurrent inflammatory processes microscopic anatomy of the cord. The epithelial thickness is that may alter the mucosal wave even in the presence of very variable, from as few as 5 to over 20 cells in thickness intraepithelial diseases, or due to subjective variation of in different places. This causes problems when dysplasia is the subtle exam interpretation, suboptimal glottic plane graded in terms of the thickness involved. There is a com- visualization or other technical limits [1–3]. Attempts to plex transition between respiratory and squamous epithelium correlate various vocal fold vibratory patterns with the and occasionally glands within the cords, whose ducts can nature of dysplasia (mild vs. severe) have also been frus- undergo metaplastic changes. Inflammation from reflux, pos- trated [2]. However, VLS, especially in combination with sible candidal infection, and irradiation from previous cancer videorecording and various bioendoscopic tools, may well treatment further complicate the histological picture. improve diagnostic accuracy in LDs and their differential The published literature shows that almost all dysplasia diagnosis from more aggressive lesions in respect to sim- grading systems differ but are effective in different ways ple videolaryngoscopic evaluation under white light (WL). [7]. The 2017 World Health Organization (WHO) system As a matter of fact, a great help in trying to distinguish [8] should is slightly more reproducible and easier for the high- from low-risk lesions (i.e., lesions with a low malig- non-specialist pathologist to apply, but may be less specific nant potential like keratosis without atypia or mild dyspla- that previous systems. Perhaps the most important point is sia) may be derived from the systematic in-office use of that the surgeon should know how their pathologist grades bioendoscopic techniques such as the Narrow Band Imag- samples and audit their transformation rate locally to calcu- ing (NBI) by Olympus or the Storz Professional Image late useful predictive values. It must be recognised that no Enhancement System (SPIES), that have been repeat- grading systems are well designed and none account for the edly demonstrated to be superior to WL endoscopy alone underlying biological process [9]. in giving a valuable insight into the biologic behaviour Although the 2017 WHO system has only been available of LDs [4]. As recently demonstrated by Stanikova and for a short time, several studies have already challenged its coworkers, both methods yield comparable outcomes in definitions and reliability [4, 10] and it should be seen as distinguishing benign/low-risk from malignant/high-risk an incremental improvement in a long process and not a lesions, with a sensitivity and specificity of 83% and 98%, perfect system. respectively, for NBI, and 86% and 96% for SPIES [5]. Taken at a simple level, the WHO system seems straight- Both techniques rely on a widely-applied classification forward to apply. The first decision is to determine whether of neoangiogenetic vascular changes that, when present, invasion is present, itself often difficult in small fragmented and poorly orientated samples. Then it is necessary to decide 13 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 1725 whether mild changes can be accounted for by reactive review published in 2012 showed that, until that moment, no benign processes rather than mild dysplasia, which can be marker or group of markers had been able to reliably indi- difficult as no distinguishing criteria are provided. If these cate the risk of progression to carcinoma. Since then some decisions are not possible, the sample may be unsuitable new data has been published which show more promising for grading. Grades may then be applied on the basis of results [11]. the thickness of the epithelium involved, proliferation and In 2014, a study was published concluding that chromo- maturation. Keratin is excluded, but whether terminally dif- some instability was associated with malignant progression ferentiated epithelium should also be included in differenti- in LD [12]. In this study, neither histopathology nor the pro- ated lesions is unclear. Provided the other criteria given in tein markers predicted progression in univariate analysis, Table 1 are met, a grade should be relatively easy to apply, although histopathological diagnosis, p53, and the apoptotic given that there are only two. Unfortunately, what to do adaptor protein FADD contributed positively to chromosome when the additional criteria listed are not met is unclear and instability in multivariate analysis. Chromosome instabil- causes most of the problems. All studies of grading systems ity was associated with malignant progression especially in have shown the importance of inter-examiner standardisa- lower grade lesions. tion and training. Applying the system with a second expe- More recently, NANOG (a master regulator of embry- rienced pathologist is essential to application of the system, onic stem cell pluripotency found to be frequently aber- to determine how such inconsistencies can be resolved and rantly expressed in a variety of cancers, including laryngeal photographic standards would be far more helpful than text SCC) has been proposed as a predictor of cancer progres- descriptions. Unfortunately, the table published in the WHO sion for LD [13]. In this study, NANOG protein expression bluebook in 2017 contains an error that will cause some con- was evaluated by immunohistochemistry using two inde- fusion in implementing the new system [4]. The amended pendent cohorts of patients with LD and correlated with correct form is shown in Table 1. clinicopathological parameters and laryngeal cancer risk. Whether carcinoma in situ should be used remains con- NANOG expression was detected by immunohistochemistry tentious and pathologists should agree with their clinicians in 49 (60%) of 82 LD, whereas expression was negligible in whether they find this additional grade useful in treatment patient-matched normal epithelia. Strong NANOG expres- decisions. In our service, the grade is not normally used, sion was found in 22 (27%) lesions and was established but may be applied after discussion by the multidiscipli- as cut-off point, showing the most robust association with nary team as it may trigger oncological treatment. Many laryngeal cancer risk (p = 0.003) superior to the histologi- published studies suggest that the additional grade provides cal classification (p = 0.320). Similar trends were obtained useful predictive information [7]. using a multicenter validation cohort of 86 patients with LD. These findings suggest that the use of NANOG expression may be of interest as a biomarker for cancer risk assessment Biomarkers in LD. In a very recent publication, mutational profiling was Due to the limited prognostic value of the histological fac- studied in LD. Using targeted next-generation sequenc- tors, biomarkers have long been investigated to try and better ing, non-synonymous mutations in six genes (PIK3CA, assess the risk of LD progressing to cancer. A systematic FGFR3, TP53, JAK3, MET, and FBXW7) were identified Table 1  Comparison of grading systems for LD, after WHO 2017 [8], with corrected levels. The published version limits low-grade dysplasia to the lower third, rather than the lower half of the epithelium Level of abnormal WHO 2005 classifica- SIN classifica- Ljubljana classifica- Amended Lju- WHO 2017 [8] maturation (WHO tion [61] tion [60] tion [59] bljana classifica- 2005) tion [58] Squamous hyperplasia Squamous hyperplasia Squamous hyperplasia Low-grade SIL Low-grade dysplasia Lower 1/3 Mild dysplasia SIN 1 Basal/parabasal hyper- plasia 1/3 to 1/2 Moderate SIN 1 or 2 Atypical hyperplasia High-grade SIL Upper 1/2–3/4 Dysplasia SIN 2 High-grade dysplasia* Full thickness Severe dysplasia Carcinoma in situ Carcinoma in situ Carcinoma in situ * A grade of carcinoma in situ may be used if a three-tiered system is preferred SIN squamous intraepithelial neoplasia, SIL intraepithelial lesion 13 1726 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 and validated by Sanger sequencing and/or qPCR. Mutations in the higher layers of the epithelium, loss of maturation, in PIK3CA and FGFR3 were detected in LD progressing to and architectural changes. These cellular abnormalities are cancer but were absent in non-progressing cases. In contrast, termed dysplasia. Dysplasia can lead to malignant transfor- mutations in JAK3, MET, and FBXW7 were found in non- mation over time [18]. Van Hulst et al. specifically studied progressing LDs but not in progressing cases. Except for the correlation between the grade of dysplasia and develop- R248W, mutations were mutually exclusive [14]. Moreover, ment of invasive laryngeal cancer in a systematic review of five of seven progression dysplasia mutations were located the medical literature [7]. They concluded rates of malignant in motif H2 of p53, whereas none of the non-progressing transformation in mild dysplasia range from 0 to 41.7%, in dysplasia mutations were. In summary, the authors proposed moderate dysplasia from 0 to 48.0%, in severe dysplasia that the mutational profile of LD can be useful for the early from 14.3 to 44.4%, and in carcinoma in situ from 11.1 to detection of patients at risk of progression. 75%. The authors found a wide variety in progression inter- The studies above reinforce the concept that LD is a vals between publications. Importantly, they found moderate complex and multifaceted entity that encompasses different dysplasia was more susceptible to malignant transformation lesions with varying ability to evolve into carcinoma. Cur- than previously thought. rent histological classifications are not powerful enough to clearly predict the risk to carcinoma evolution. Recently, different studies have defined the possible utility of different Treatment biomarkers: chromosome instability, NANOG, and muta- tional profiles have been able to predict this risk more accu- The variety of interventions reported and the lack of clarity rately and have demonstrated their practical utility. regarding interventions and follow-up regimes have made the management of dysplasia very controversial [19]. Most of our current knowledge on the treatment of LD has been Natural course, recurrence rates, progression derived from publications with a focus on early stage glot- to infiltrating carcinoma tic carcinomas. These results are frequently applied to LD patients, presuming that early invasive cancer and LD will Isenberg et al. evaluated 2188 biopsies from laryngeal leu- react similarly to different treatments. Although this assump- koplakia and identified mild or moderate dysplasia in 33.5% tion may not be entirely justified, most of our understanding and severe dysplasia or carcinoma in situ in 15.2% [15]. of LD treatment will have to rely on data obtained from early Only histological evaluation could settle a final diagnosis as stage infiltrating cancer, until more relevant and specific data to the presence or absence of dysplasia. Laryngoscopy alone for LD become available. or in combination with VLS could not reliably allow for a The great majority of patients treated with LD have been final diagnosis. During follow-up (1–233 months) infiltrat- treated with surgical methods. However, a minority will ing laryngeal SCC was diagnosed in 3.7% of patients with be treated with radiotherapy, for different reasons (usually, no dysplastic lesions initially, 10.1% of patients with mild repeatedly recurring dysplasia) [16, 19, 20]. In most Euro- or moderate dysplasia, and 18.1% of patients with severe pean institutions, transoral laser microsurgery is now con- dysplasia. Karatayli-Ozgursoy et al. described a 15.9% trans- sidered the standard of care. formation rate to invasive carcinoma for mild dysplasia in a 20-year single institution review [16]. A 12.1% rate for moderate and a 23.4% rate for severe dysplasia were noted. Transoral laser microsurgery Transformation occurred in 48.6% of carcinoma in situ. In contrast, the risk for developing laryngeal SCC out of While different types of lasers have been used in laryn- LD showed no statistical correlation to the initial dyspla- geal surgery, the C­ O2 is still the undisputed workhorse for sia grade in a recent review on 70 patients [17]. On aver- microlaryngeal surgery. Owing to its frequent use in tumour age, malignant conversion took 127 weeks, 117 for mild surgery, it is available in the Otolaryngology—Head and dysplasia, 135 weeks for moderate dysplasia, and 82 weeks Neck Surgery Departments of most hospitals. The C ­ O2 laser for severe dysplasia. The authors concluded that patients meets the requirements for use for most benign and malig- with LD are an inhomogeneous group and, as already stated nant laryngeal lesions. In the American literature, angiolytic above, the grade of LD alone seems to be an insufficient lasers have repeatedly been recommended for laryngeal sur- prognostic factor for the development of laryngeal cancer. gery [21]. However, they have not found wider acceptance It is presumed that LD have a higher rate of malignant in Europe so far and large comparison studies demonstrat- transformation compared to normal epithelium. The under- ing their superiority are still lacking. Additional equipment lying concept is that consecutive mutations in normal epi- is available for the optimal delivery of laser energy to the thelial cells lead to nuclear abnormalities, such as mitoses operative site, including devices for focusing the surgical 13 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 1727 beam to an extremely small spot for precision cutting. Scan- surgical margins, granulomas mimicking persistent/recurrent ner systems are also available for coagulating predefined lesions, web formation or other post-excisional abnormal mucosal areas while preserving the underlying tissue. Tran- tissue growth at the level of the primary resection site poten- soral laser microsurgery (TLM) has become the most widely tially impacting on functional outcomes in spite of appropri- used therapeutic approach for laryngeal keratosis, dysplasia, ate medical treatment and voice therapy, or involvement of carcinoma in situ, and early stage laryngeal squamous cell certain laryngeal subsites (anterior commissure, ventricle, carcinoma [22, 23]. The results have been documented in subglottis) [33]. The timing of SLM is still matter of dis- numerous retrospective and some prospective case series cussion, being variably reported in the literature as ranging [24]. It is generally used as a single modality treatment with from 1 to 8 months after primary surgery. The same is true curative intent and results in terms of local control rates are for the need of a third- or further-look microlaryngoscopy. comparable to both open surgery and radiation therapy, with However, both prospective and retrospective series clearly significantly less expenses. Immediate as well as long-term demonstrate a benefit in terms of earlier detection of persis- morbidities were also reduced when compared to both of tence/recurrence (even in the presence of previous micro- them. scopically free surgical margins). This is especially true in Laryngeal exposure is one of the most important factors case of anterior commissure involvement, and apparently influencing TLM resection of glottic cancer within safe normal larynx at fiberoptic evaluation [33, 34]. Moreover, surgical margins [25]. Anterior commissure involvement is an adjunctive advantage of SLM is the possibility to confirm associated with an increased risk of recurrence and possi- the benign nature of symptomatic granulomas or webs and to ble deterioration of the voice quality [26–28]. Type 1 or 2 promote their prompt surgical management with a functional resections (according to the European Laryngological Soci- gain in terms of voice and/or airway patency. ety [ELS] classification of different transoral cordectomies [29]) are usually performed. In their literature review on the management of LD, Sadri Radiotherapy et al. found a pooled overall local control rate of 81% for TLM, as opposed to 91% for RT in treatment of severe dys- Radiotherapy (RT) has also been employed in the manage- plasia and carcinoma in situ of the larynx [19]. In contrast, ment of LD. It has been used mainly on patients with high Peretti et al. in their retrospective single institution review grade or severe dysplasia but its use and indications have of TLM by ­CO2 laser in 71 patients reported a 5-year dis- remained controversial. It should be considered in patients ease-specific survival, local control with laser alone, loco- who have multiple sites or field change in which surgical regional, regional control, and organ preservation rates of excision could lead to significant scarring and sequelae 100%, 93.4%, 100%, 100%, and 98.5%, respectively [30]. [35]. The main indications for RT in LD are summarised From these data, it is obvious that the results of ­CO2 laser in Table 2. surgery have improved over time. It is also obvious that the RT regimes used are similar to those applied for T1a results of any given surgical procedure will largely depend carcinomas of the larynx [19]. The risks of progression to on the experience of the individual surgeon and the disease- invasive carcinoma in patients treated by RT for severe dys- specific workload of the institution delivering specialised plasia or carcinoma in situ has been reported to be around services in the head and neck oncology and laryngology 12% [19]. Local control rates (LCR) range from 59 to 100%. fields. Most recent series offer LCR between 79 and 98% [36, 37]. While RT gave higher local control rates than laser sur- Patients most at risk are those who continue to smoke and gery when used as the initial treatment in the past, TLM drink and those who have lesions in the anterior commis- can be used repeatedly in case of local recurrence, and the sure [38]. final laryngectomy-free survival rates seem to be higher with The main rationale for many authors for the use of RT initial laser surgery as compared to radiation therapy (obvi- has been the voice outcomes after such treatment. Persistent ously a consequence of better re-treatment options in case dysphonia has been found more frequently following TLM of new dysplastic manifestations after primary treatment) than after RT. Also on stroboscopic examination, incomplete [31, 32]. Although these findings are based primarily on the results for early infiltrating carcinoma, it is reasonable Table 2  Indications for the to believe that the same outcome can be expected for high- management of LD with 1 Multiple recurrences grade dysplasia and carcinoma in situ. radiotherapy 2 Persistent smoking and inability to cease habit The need for a systematic second look microlaryngoscopy 3 High anaesthetic risk (SLM) under general anaesthesia at 6–12 weeks after the 4 Patient’s preference first procedure remains highly controversial. It can be dic- 5 Wide-field multiple lesions tated by uncertain (close or altered for iatrogenic artefacts) 13 1728 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 glottic closure and diminution or lack of vibration of the To our knowledge, there are not specific publications on operated vocal fold have been frequently observed follow- results with open surgery in the particular setting of LD in ing TLM. However, when maximum phonation time, mean the last 25 years due do the development of transoral surgery airflow rate, fundamental frequency range of phonation, and but the results can be expected similar to those obtained by intensity range of phonation have been analysed side to side transoral approach. no differences have been found [39]. RT is now considered a treatment option for severe or high-grade LD in a lim- ited subset of patients with repeated recurrences, wide-field Medical treatment and prevention lesions, and high anaesthetic risks. Smoking cessation is clearly the most important issue in the prevention of LD and of recurrences following initial treatment [41]. There is a great, unmet clinical need for treat- Conventional microlaryngoscopic surgery ments of LD that provide nonsurgical options to patients and treat the entire epithelial field. A treatment agent that con- Apart from laser surgery, resection can be accomplished vincingly and substantially reduces LD burden might also with sharp or cutting instruments (knives, scissors) or elec- decrease invasive cancer risk and the subsequent need for trosurgical instruments. Microinstruments are available for surgical interventions. microsurgery of the larynx and trachea, but these require For the time being, chemotherapy and immunotherapy handling by an experienced surgeon. Also, the same time, are not an issue in LD. For the vast majority of all patients, the division of tissues with cutting instruments always there are currently no meaningful medical approaches to the involves capillaries and small arterial or venous vessels, disease. In patients with LD related to laryngeal papilloma- leading to diffuse bleeding at the surgical site. Although the tosis, HPV vaccination, treatment with Cidofovir [42] and volume of blood loss in the larynx is usually not an issue, the bevacizumab [43] may be considered on an individual basis. bleeding nevertheless obscures the operative field and makes Recently, both aspirin [44–46] and metformin [47–49] it difficult to assess the progress of the resection, often slow- have emerged as promising medications for the prevention of ing the procedure down. This type of bleeding is particularly dysplasia formation in man. A small case series has reported troublesome in the situation of narrow margin microsurgery on promising observations in nondiabetic patients treated such as is the case in LD. No reliable data on the results with metformin for recurrent LD lesions [49]. Clinical of cold instrument surgery for LD have been published in long-term results, however, are not available to date. Folate recent years. Sadri et al. noted a pooled local control rate of deficiency may be considered to be a factor predisposing 77% in their literature review [19]. A reasonable alternative to precancerous lesions, and dietary folate supplementation to TLM may be the use of microdissection electrodes [40]. may prevent and reduce the emergence of cancer [50] However, again there is a lack of large series comparable to those reported for TLM. Follow‑up Open surgery It is accepted that the follow-up for patients treated for LD is an essential part of their care. The reasons of post-treatment The transoral approach is the main way to treat surgically follow-up in the setting of laryngeal cancer in general have LD. In particular cases; however, open surgery still can have been previously reported by Simo et al. [51]. However, con- a role. The situation and the discussion is pretty similar to troversy still exists in how these aims should be achieved, that in T1a vocal fold carcinoma and the main indication is while increasing efforts are being made to rationalize the when there is not an adequate exposure for transoral surgery, structure and timing of follow-up clinics to minimize pres- for example in LD located at the level of the posterior com- sure on the health care systems ensuring appropriate stand- missure. Open surgery has the advantage of excellent access ard of care. to the lesions and the resection can be adapted according to The length of follow-up has been described to be very the dysplasia size. variable. Cosway and Weller recently reported an evidence- Currently, the only indication for open partial resection based chart to aid clinicians with the follow-up of LD after in LD is the inadequate exposure by a conventional transoral excision [35, 52]. They recommended a minimum follow-up approach. The surgical approach is usually performed via a of 6 months for patients with low-grade or mild-to-moderate midline thyrotomy. If needed, to have a better vision of the dysplasia with no risk factors, while a follow-up for 5 years posterior larynx, the orotracheal intubation can be temporar- in those with high-risk or severe dysplasia [35]. Follow-up ily placed though the inferior part of the thyrotomy. for more than 5 years can be justified in those who have 13 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 1729 progressed or recurred after appropriate treatment, and have proven to be of great value for the prevention of pro- for those who continue to smoke and drink heavily. Fear gression and recurrence of LD. However, future behavioural of recurrence is prevalent in cancer patients and continued change interventions should be conducted within popula- attendance at clinic helps to mitigate this issue [53]. tions with LD [55]. At present, there is no evidence that a high frequency Evidence to support follow-up for early detection of LD of follow-up visits confers any benefit in terms of reducing progression or recurrence is lacking, although there is a morbidity and mortality. As a consequence, the majority of common belief that follow-up clinics have inherent value in clinicians support a follow-up schedule with decreasing fre- detecting morphological and progression changes. However, quency overtime. In the first 2 years, when the risk of loco- with the current socio-economic climate and with health regional recurrence is high, the frequency should remain care systems reducing costs, there is a real risk that long- high and be followed by a gradual decrease in the second term follow-up for patients with LD or even laryngeal cancer year. Therefore, the follow-up in the first 2 years should be may be seriously affected in the near future. Therefore, it in between 4 and 8 weeks and from 3 to 6 months thereafter, may be needed that health care professionals provide risk- especially when considering patients managed for a high- assessment strategies to rationalize potential unnecessary grade LD [35] follow-up, for example subdividing patients with LD into At present, most patients tend to be followed-up by their low- and high-risk of developing progression to invasive treating clinicians. While in most European countries follow- carcinoma. It is believed that this categorization could help up is usually performed by Otolaryngologists, in the United to determine how and for how long patients should be fol- Kingdom, introduction of clinical nurse specialists and the lowed for. It would also help to establish which screening key worker role in management of patients with HNC, has tests might be needed to detect recurrence or progression to become vital to open lines of communication between the invasive cancer [22]. patient and family, their careers, their general practitioner and the clinical team should any problems arise [51]. This could also be also extended to the management of LD. New technologies in the follow‑up Traditionally, clinical assessment has been the most important aspect of the follow-up in patients treated for LD. The ideal clinical investigation tool for patients for follow- The clinical evaluation should include inspection of the up after treatment for LD is flexible nasopharyngolaryngo- larynx, employing the use of transoral rigid enedoscope or scope. Videorecording using high resolution flexible vid- transnasal video- or fibroscopy and palpation of the neck. eoendoscopes is paramount for accurate documentation, By focusing on vocal fold vibration during phonation, using archiving of images, minimising inter-observer discrepan- VLS, the laryngologist can contribute considerably to the cies and allowing comparisons during the follow-up period. diagnosis of persistent/recurrent LD while differentiating Narrow Band Imaging (NBI), with associated high definition this from attended scar tissues or other iatrogenic sequelae. television technology, can be an excellent adjunctive imag- The diagnostic accuracy of laryngeal imaging in general is ing tool. This is owed to its specific capability to selectively up to 68%. Particular diagnoses, however, are more consist- address superficial persistent changes/recurrences or second ently identified; cancer, for example, was much more accu- primary tumors by enhancing their pathognomonic neoan- rately identified on laryngoscopy (100%) and VLS (100%) giogenetic pattern. It has been reported that its use can detect than on history and physical examination alone (33%) [54]. 18% more true positive laryngeal cancer lesions than con- Moreover, photo- and video documentation allows to com- ventional white light (WL) endoscopy. This has been also pare changes, playing an important role also from an aca- demonstrated after RT, due to the high accuracy (98%) of demic and patient’s educational point of view. There is little NBI in differentiating between neoplastic disease and post- evidence that CT, MR, and PET scanning may play a role in RT inflammatory and/or cicatricial changes [56, 57]. the follow-up of LD unless there is a suspect of progression A recent systematic review by Paleri et al. demonstrated to invasive carcinoma. that with a histological diagnosis of mild dysplasia which Patient’s education is a key factor in the management of show no abnormalities on NBI, the post-test probability of LD and its incorporation into follow-up symptom evaluation malignancy is 2.3% compared to 10.3 with conventional WL is to be recommended. As a matter of fact, the understand- imaging. For severe dysplasia, similar post-test probabilities ing of the potential symptoms and signs of recurrence is a after NBI and WL are 8.0 and 29.7%, respectively. Post-test key factor in early diagnosis and patients should be aware probabilities in this setting indicate the chance of missing of these so they can seek medical attention as soon as they malignancy following NBI or WL in patients who undergo perceive any subtle modification of their voice, lasting more no further intervention [34]. than 2–3 weeks. Patient’s education should also include Recently a specific Image Enhancement System (SPIES) tobacco smoking and alcohol cessation programmes, which [5] has been accessible to clinicians; however, no robust data 13 1730 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 have been reported yet. However, the possible advantages of included in the article’s Creative Commons licence, unless indicated its use in detection of recurrent/persistent or progressing LD otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not may be similar to those already described for NBI, due to permitted by statutory regulation or exceeds the permitted use, you will the overlapping information obtainable with such a device. need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. Conclusion References The diagnosis of LD largely relies on endoscopic procedures 1. van Balkum M, Buijs B, Donselaar EJ, Erkelens DC, Goulin Lippi Fernandes E, Wegner I, Grolman W, Janssen LM (2017) System- and on histopathology. Diagnostic efficiency of endoscopy atic review of the diagnostic value of laryngeal stroboscopy in may be improved using VLS and bioendoscopic tools such excluding early glottic carcinoma. Clin Otolaryngol 42(1):123– as NBI and SPIES. Current histological classifications are 130. https​://doi.org/10.1111/coa.12678​ not powerful enough to clearly predict the risk to carcinoma 2. Djukic V, Milovanovic J, Jotic AD, Vukasinovic M (2014) Stro- boscopy in detection of laryngeal dysplasia effectiveness and limi- evolution and technical issues such as sampling error, varia- tations. J Voice 28(2):e213–e221. https​://doi.org/10.1016/j.jvoic​ tion in epithelial thickness and inflammation hamper patho- e.2013.07.006 logical examination. Almost all dysplasia grading systems 3. Peretti G, Piazza C, Berlucchi M, Cappiello J, Giudice M, Nicolai differ but are effective in different ways. The 2017 WHO sys- P (2003) Pre- and intraoperative assessment of mid-cord eryth- roleukoplakias: a prospective study on 52 patients. Eur Arch tem should prove to be an improvement as it is slightly more Otorhinolaryngol 260(10):525–528. https:​ //doi.org/10.1007/s0040​ reproducible and easier for the non-specialist pathologist to 5-003-0584-0 apply. To optimize treatment decisions, surgeons should 4. Mehlum CS, Larsen SR, Kiss K, Groentved AM, Kjaergaard T, know how their pathologist grades samples and preferably Moller S, Godballe C (2018) Laryngeal precursor lesions: inter- rater and intrarater reliability of histopathological assessment. audit their transformation rates locally. Whether carcinoma Laryngoscope 128(10):2375–2379. https​: //doi.org/10.1002/ in situ should be used as a classification remains contentious lary.27228​ and pathologists should agree with their clinicians whether 5. Stanikova L, Walderova R, Jancatova D, Formanek M, Zelenik they find this additional grade useful in treatment decisions. K, Kominek P (2018) Comparison of narrow band imaging and the Storz Professional Image Enhancement System for detection Recently, different studies have defined the possible util- of laryngeal and hypopharyngeal pathologies. Eur Arch Otorhi- ity of different biomarkers in risk classification. The main nolaryngol 275(7):1819–1825. https​://doi.org/10.1007/s0040​ treatment modality for LD is TLM. RT may be indicated in 5-018-4987-3 specific circumstances such as multiple recurrence or wide- 6. Arens C, Piazza C, Andrea M, Dikkers FG, Tjon Pian Gi RE, Voigt-Zimmermann S, Peretti G (2016) Proposal for a descrip- field lesions. Medical treatment currently does not have a tive guideline of vascular changes in lesions of the vocal folds by significant role in the management of LD. Follow-up for the committee on endoscopic laryngeal imaging of the European patients treated with LD is a fundamental part of their care Laryngological Society. Eur Arch Otorhinolaryngol 273(5):1207– and investigations may be supported by the same techniques 1214. https​://doi.org/10.1007/s0040​5-015-3851-y 7. van Hulst AM, Kroon W, van der Linden ES, Nagtzaam L, Otten- used during diagnosis (VLS and NBI/SPIES). hof SR, Wegner I, Gunning AC, Grolman W, Braunius W (2016) Grade of dysplasia and malignant transformation in adults with premalignant laryngeal lesions. Head Neck 38(1):E2284–2290. Funding No funding was obtained for this review. https​://doi.org/10.1002/hed.24185​ 8. El-Naggar AK, JKC C, JR G, T T, PJ S (2017) WHO 2017 Classi- fication of Tumours, 4th Edition, Volume 9 Head and Neck. WHO Compliance with ethical standards Classification of Tumours. WHO Press, World Health Organiza- tion, Geneva, Switzerland Conflict of interest All authors declare that he/she has no conflict of 9. Cross SS (1998) Grading and scoring in histopathology. Histopa- interest. thology 33(2):99–106 10. Cho KJ, Song JS (2018) Recent changes of classification for squa- Ethical approval This review did not include individual patient data mous intraepithelial lesions of the head and neck. Arch Pathol Lab and was conducted in accordance with the ethical standards of the insti- Med 142(7):829–832. https:​ //doi.org/10.5858/arpa.2017-0438-RA tutional and/or national research committee and with the 1964 Helsinki 11. Rodrigo JP, Garcia-Pedrero JM, Suarez C, Takes RP, Thomp- declaration and its later amendments or comparable ethical standards. son LD, Slootweg PJ, Woolgar JA, Westra WH, Brakenhoff RH, Rinaldo A, Devaney KO, Williams MD, Gnepp DR, Ferlito A (2012) Biomarkers predicting malignant progression of laryn- Open Access This article is licensed under a Creative Commons Attri- geal epithelial precursor lesions: a systematic review. Eur Arch bution 4.0 International License, which permits use, sharing, adapta- Otorhinolaryngol 269(4):1073–1083. https​://doi.org/10.1007/ tion, distribution and reproduction in any medium or format, as long s0040​5-011-1831-4 as you give appropriate credit to the original author(s) and the source, 12. Bergshoeff VE, Van der Heijden SJ, Haesevoets A, Litjens SG, provide a link to the Creative Commons licence, and indicate if changes Bot FJ, Voogd AC, Chenault MN, Hopman AH, Schuuring E, were made. The images or other third party material in this article are Van der Wal JM, Manni JJ, Ramaekers FC, Kremer B, Speel EJ 13 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 1731 (2014) Chromosome instability predicts progression of premalig- involving the anterior commissure: voice and oncologic out- nant lesions of the larynx. Pathology 46(3):216–224. https​://doi. comes. Eur Arch Otorhinolaryngol 272(2):413–418. https​://doi. org/10.1097/PAT.00000​00000​00006​8 org/10.1007/s0040​5-014-3368-9 13. Rodrigo JP, Villaronga MA, Menendez ST, Hermida-Prado F, 28. Rucci L, Romagnoli P, Scala J (2010) CO(2) laser therapy in Quer M, Vilaseca I, Allonca E, Pedregal Mallo D, Astudillo A, Tis and T1 glottic cancer: indications and results. Head Neck Garcia-Pedrero JM (2017) A novel role for nanog as an early can- 32(3):392–398. https​://doi.org/10.1002/hed.21200​ cer risk marker in patients with laryngeal precancerous lesions. 29. Remacle M, Eckel HE, Antonelli A, Brasnu D, Chevalier D, Frie- Sci Rep 7(1):11110. https:​ //doi.org/10.1038/s41598​ -017-11709-​ 9 drich G, Olofsson J, Rudert HH, Thumfart W, de Vincentiis M, 14. Manterola L, Aguirre P, Larrea E, Arestin M, Gaafar A, Elorriaga Wustrow TP (2000) Endoscopic cordectomy. A proposal for a K, Goicoechea I, Armesto M, Fernandez-Mercado M, Zabalza I, classification by the Working Committee European Laryngologi- Lopez-Duque JC, Larruskain E, Sistiaga JA, Landa M, Zabala A, cal Society. Eur Arch Otorhinolaryngol 257(4):227–231 Santaolalla F, Municio JA, Ispizua A, Garcia-Pedrero JM, Rodrigo 30. Peretti G, Piazza C, Cocco D, De Benedetto L, Del Bon F, Redaelli JP, Lawrie CH (2018) Mutational profiling can identify laryngeal De Zinis LO, Nicolai P (2010) Transoral CO(2) laser treatment for dysplasia at risk of progression to invasive carcinoma. Sci Rep T(is)-T(3) glottic cancer: the University of Brescia experience on 8(1):6613. https​://doi.org/10.1038/s4159​8-018-24780​-7 595 patients. Head Neck 32(8):977–983. https​://doi.org/10.1002/ 15. Isenberg JS, Crozier DL, Dailey SH (2008) Institutional and com- hed.21278​ prehensive review of laryngeal leukoplakia. Ann Otol Rhinol Lar- 31. Mahler V, Boysen M, Brondbo K (2010) Radiotherapy or CO(2) yngol 117(1):74–79. https​://doi.org/10.1177/00034​89408​11700​ laser surgery as treatment of T(1a) glottic carcinoma? Eur Arch 114 Otorhinolaryngol 267(5):743–750. https​://doi.org/10.1007/s0040​ 16. Karatayli-Ozgursoy S, Pacheco-Lopez P, Hillel AT, Best SR, 5-009-1135-0 Bishop JA, Akst LM (2015) Laryngeal dysplasia, demographics, 32. Schrijvers ML, van Riel EL, Langendijk JA, Dikkers FG, Schuur- and treatment: a single-institution, 20-year review. JAMA Otolar- ing E, van der Wal JE, van der Laan BF (2009) Higher laryngeal yngol Head Neck Surg 141(4):313–318. https​://doi.org/10.1001/ preservation rate after C ­ O2 laser surgery compared with radio- jamao​to.2014.3736 therapy in T1a glottic laryngeal carcinoma. Head Neck 31(6):759– 17. Luers JC, Sircar K, Drebber U, Beutner D (2014) The impact of 764. https​://doi.org/10.1002/hed.21027​ laryngeal dysplasia on the development of laryngeal squamous 33. Preuss SF, Cramer K, Drebber U, Klussmann JP, Eckel HE, Gun- cell carcinoma. Eur Arch Otorhinolaryngol 271(3):539–545. https​ tinas-Lichius O (2009) Second-look microlaryngoscopy to detect ://doi.org/10.1007/s0040​5-013-2670-2 residual carcinoma in patients after laser surgery for T1 and T2 18. Argiris A, Karamouzis MV, Raben D, Ferris RL (2008) Head laryngeal cancer. Acta Otolaryngol 129(8):881–885. https​://doi. and neck cancer. Lancet 371(9625):1695–1709. https​: //doi. org/10.1080/00016​48080​24417​39 org/10.1016/S0140​-6736(08)60728​-X 34. Paleri V, Sawant R, Mehanna H, Ainsworth H, Stocken D (2018) 19. Sadri M, McMahon J, Parker A (2006) Management of laryngeal Laryngeal dysplasia and narrow band imaging: secondary analy- dysplasia: a review. Eur Arch Otorhinolaryngol 263(9):843–852. sis of published data supports the role in patient follow-up. Clin https​://doi.org/10.1007/s0040​5-006-0078-y Otolaryngol 43(6):1439–1442. https:​ //doi.org/10.1111/coa.13182​ 20. Gale N, Gnepp DR, Poljak M, Strojan P, Cardesa A, Helliwell 35. Cosway B, Paleri V (2015) Laryngeal dysplasia: an evidence- T, Sifrer R, Volavsek M, Sandison A, Zidar N (2016) Laryngeal based flowchart to guide management and follow up. J Laryngol squamous intraepithelial lesions: an updated review on etiology, Otol 129(6):598–599. https​://doi.org/10.1017/S0022​21511​50008​ classification, molecular changes, and treatment. Adv Anat Pathol 33 23(2):84–91. https​://doi.org/10.1097/PAP.00000​00000​00010​6 36. Le QT, Takamiya R, Shu HK, Smitt M, Singer M, Terris DJ, Fee 21. Zeitels SM (2016) Glottic cancer: a metamorphosing dis- WE, Goffinet DR, Fu KK (2000) Treatment results of carcinoma ease. Ann Otol Rhinol Laryngol 125(6):452–456. https​://doi. in situ of the glottis: an analysis of 82 cases. Arch Otolaryngol org/10.1177/00034​89415​61917​7 Head Neck Surg 126(11):1305–1312 22. Paleri V, Mackenzie K, Wight RG, Mehanna H, Pracy P, Bradley 37. Spayne JA, Warde P, O’Sullivan B, Payne D, Liu FF, Waldron J, PJ, Head E-U, Neck G (2009) Management of laryngeal dysplasia Gullane PJ, Cummings BJ (2001) Carcinoma-in-situ of the glottic in the United Kingdom: a web-based questionnaire survey of cur- larynx: results of treatment with radiation therapy. Int J Radiat rent practice. Clin Otolaryngol 34(4):385–389. https​://doi.org/10. Oncol Biol Phys 49(5):1235–1238 1111/j.1749-4486.2009.01938​.x 38. Sllamniku B, Bauer W, Painter C, Sessions D (1989) The trans- 23. Peretti G, Piazza C, Ansarin M, De Benedetto L, Cocco D, Cat- formation of laryngeal keratosis into invasive carcinoma. Am J taneo A, Nicolai P, Chiesa F (2010) Transoral ­CO2 laser microsur- Otolaryngol 10(1):42–54 gery for Tis-T3 supraglottic squamous cell carcinomas. Eur Arch 39. Verdonck-de Leeuw IM, Keus RB, Hilgers FJ, Koopmans-van Otorhinolaryngol 267(11):1735–1742. https​://doi.org/10.1007/ Beinum FJ, Greven AJ, de Jong JM, Vreeburg G, Bartelink H s0040​5-010-1284-1 (1999) Consequences of voice impairment in daily life for patients 24. Huang G, Luo M, Zhang J, Liu H (2017) Laser surgery versus following radiotherapy for early glottic cancer: voice quality, vocal radiotherapy for T1a glottic carcinoma: a meta-analysis of onco- function, and vocal performance. Int J Radiat Oncol Biol Phys logic outcomes. Acta Otolaryngol 137(11):1204–1209. https​:// 44(5):1071–1078 doi.org/10.1080/00016​489.2017.13537​06 40. Basterra J, Zapater E, Moreno R, Hernandez R (2006) Electrosur- 25. Damm M, Sittel C, Streppel M, Eckel HE (2000) Transoral C ­ O2 gical endoscopic cordectomy with microdissection electrodes: a laser for surgical management of glottic carcinoma in situ. Laryn- comparative study with C ­ O2 laser. J Laryngol Otol 120(8):661– goscope 110(7):1215–1221. https​://doi.org/10.1097/00005​537- 664. https​://doi.org/10.1017/S0022​21510​60013​68 20000​7000-00028​ 41. Al-Mamgani A, van Rooij PH, Mehilal R, Verduijn GM, Tans L, 26. Hoffmann C, Cornu N, Hans S, Sadoughi B, Badoual C, Brasnu Kwa SL (2014) Radiotherapy for T1a glottic cancer: the influ- D (2016) Early glottic cancer involving the anterior commissure ence of smoking cessation and fractionation schedule of radio- treated by transoral laser cordectomy. Laryngoscope 126(8):1817– therapy. Eur Arch Otorhinolaryngol 271(1):125–132. https​://doi. 1822. https​://doi.org/10.1002/lary.25757​ org/10.1007/s0040​5-013-2608-8 27. Mendelsohn AH, Kiagiadaki D, Lawson G, Remacle M (2015) 42. Tjon Pian Gi RE, Dietz A, Djukic V, Eckel HE, Friedrich G, CO2 laser cordectomy for glottic squamous cell carcinoma Golusinski W, Hantzakos A, Lawson G, Remacle M, Rihkanen 13 1732 European Archives of Oto-Rhino-Laryngology (2021) 278:1723–1732 H, Dikkers FG (2012) Treatment of recurrent respiratory papillo- meta-analysis. Clin Otolaryngol 35(5):364–372. https​://doi.org/1 matosis and adverse reactions following off-label use of cidofovir 0.1111/j.1749-4486.2010.02181​.x (Vistide(R)). Eur Arch Otorhinolaryngol 269(2):361–362. https:​ // 53. van Agthoven M, van Ineveld BM, de Boer MF, Leemans CR, doi.org/10.1007/s0040​5-011-1804-7 Knegt PP, Snow GB, Uyl-de Groot CA (2001) The costs of head 43. Zur KB, Fox E (2017) Bevacizumab chemotherapy for man- and neck oncology: primary tumours, recurrent tumours and agement of pulmonary and laryngotracheal papillomatosis in a long-term follow-up. Eur J Cancer 37(17):2204–2211. https​:// child. Laryngoscope 127(7):1538–1542. https​://doi.org/10.1002/ doi.org/10.1016/s0959​-8049(01)00292​-1 lary.26450​ 54. Paul BC, Chen S, Sridharan S, Fang Y, Amin MR, Branski RC 44. Jacobs EJ, Newton CC, Gapstur SM, Thun MJ (2012) Daily aspi- (2013) Diagnostic accuracy of history, laryngoscopy, and stro- rin use and cancer mortality in a large US cohort. J Natl Cancer boscopy. Laryngoscope 123(1):215–219. https​://doi.org/10.1002/ Inst 104(16):1208–1217. https​://doi.org/10.1093/jnci/djs31​8 lary.23630​ 45. Algra AM, Rothwell PM (2012) Effects of regular aspirin on long- 55. Kothari P, Trinidade A, Hewitt RJD, Singh A, O’Flynn P (2011) term cancer incidence and metastasis: a systematic comparison The follow-up of patients with head and neck cancer: an analysis of evidence from observational studies versus randomised tri- of 1039 patients. Eur Arch Otorhinolaryngol 268(8):1191–1200. als. Lancet Oncol 13(5):518–527. https​://doi.org/10.1016/S1470​ https​://doi.org/10.1007/s0040​5-010-1461-2 -2045(12)70112​-2 56. Piazza C, Del Bon F, Peretti G, Nicolai P (2012) Narrow band 46. Rothwell PM, Price JF, Fowkes FG, Zanchetti A, Roncaglioni imaging in endoscopic evaluation of the larynx. Curr Opin Oto- MC, Tognoni G, Lee R, Belch JF, Wilson M, Mehta Z, Meade laryngol Head Neck Surg 20(6):472–476. https:​ //doi.org/10.1097/ TW (2012) Short-term effects of daily aspirin on cancer inci- MOO.0b013​e3283​5908a​c dence, mortality, and non-vascular death: analysis of the time 57. Piazza C, Cocco D, De Benedetto L, Del Bon F, Nicolai P, Peretti course of risks and benefits in 51 randomised controlled trials. G (2010) Narrow band imaging and high definition television in Lancet 379(9826):1602–1612. https​://doi.org/10.1016/S0140​ the assessment of laryngeal cancer: a prospective study on 279 -6736(11)61720​-0 patients. Eur Arch Otorhinolaryngol 267(3):409–414. https​://doi. 47. Curry J, Johnson J, Tassone P, Vidal MD, Menezes DW, Sprandio org/10.1007/s0040​5-009-1121-6 J, Mollaee M, Cotzia P, Birbe R, Lin Z, Gill K, Duddy E, Zhan 58. Gale N, Blagus R, El-Mofty SK, Helliwell T, Prasad ML, Sandi- T, Leiby B, Reyzer M, Cognetti D, Luginbuhl A, Tuluc M, Mar- son A, Volavsek M, Wenig BM, Zidar N, Cardesa A (2014) Evalu- tinez-Outschoorn U (2017) Metformin effects on head and neck ation of a new grading system for laryngeal squamous intraepi- squamous carcinoma microenvironment: Window of opportunity thelial lesions–a proposed unified classification. Histopathology trial. Laryngoscope 127(8):1808–1815. https​://doi.org/10.1002/ 65(4):456–464. https​://doi.org/10.1111/his.12427​ lary.26489​ 59. Gale N, Kambic V, Michaels L, Cardesa A, Hellquist H, Zidar N, 48. Rego DF, Elias ST, Amato AA, Canto GL, Guerra EN (2017) Poljak M (2000) The Ljubljana classification: a practical strategy Anti-tumor effects of metformin on head and neck carcinoma cell for the diagnosis of laryngeal precancerous lesions. Adv Anat lines: a systematic review. Oncol Lett 13(2):554–566. https​://doi. Pathol 7(4):240–251. https​://doi.org/10.1097/00125​480-20000​ org/10.3892/ol.2016.5526 7040-00006​ 49. Lerner MZ, Mor N, Paek H, Blitzer A, Strome M (2017) Met- 60. Crissman JD, Sakr WA (2001) Squamous neoplasia of the upper formin prevents the progression of dysplastic mucosa of the head aerodigestive stract: Intraepithelial and invasive squamous cell and neck to carcinoma in nondiabetic patients. Ann Otol Rhinol carcinoma. In: Pilch BZ (ed) Head and Neck Surgical Pathology. Laryngol 126(4):340–343. https​://doi.org/10.1177/00034​89416​ Lippincott Williams & Wilkins, Philadelphia, PA, pp 34–52 68847​8 61. Gale N, Pilch BZ, Sidransky D, El-Naggar AK, Westra W, Cal- 50. Mesolella M, Iengo M, Testa D, Ricciardiello F, Iorio B (2017) ifano J et al (2005) Tumors of the oral cavity and oropharynx Chemoprevention using folic acid for dysplastic lesions of the (Epithelial precursus lesions). In: Barnes L, Eveson JW, Reich- larynx. Mol Clin Oncol 7(5):843–846. https​://doi.org/10.3892/ art P, Lyon SD (eds) World Health Organization classification of mco.2017.1424 tumours, pathology and genetics. Head and Neck tumours. Inter- 51. Simo R, Bradley P, Chevalier D, Dikkers F, Eckel H, Matar N, national Agency for Research on Cancer (IARC). IARC Press, pp Peretti G, Piazza C, Remacle M, Quer M (2014) European Lar- 177–179 yngological Society: ELS recommendations for the follow-up of patients treated for laryngeal cancer. Eur Arch Otorhinolaryngol Publisher’s Note Springer Nature remains neutral with regard to 271(9):2469–2479. https​://doi.org/10.1007/s0040​5-014-2966-x jurisdictional claims in published maps and institutional affiliations. 52. Weller MD, Nankivell PC, McConkey C, Paleri V, Mehanna HM (2010) The risk and interval to malignancy of patients with laryngeal dysplasia; a systematic review of case series and 13