Discussion
Laryngeal malignancies almost always arise from the surface epithelium and thus, more than 95% are squamous cell carcinoma.[5] Similarly, our results are consistent with the literature with an incidence of 90% conventional SCC, 7.3% vSCC and 2.7 non-SCC laryngeal malignancies.
Verrucous carcinoma is the most common variant of SCC both in the literature and in our series. In a review published in 2016, the incidence of laryngeal verrucous SCC was reported to be 1-3.4%.[6] In our study, the incidence of verrucous carcinoma within the entire laryngeal malignancies is calculated as 2.3% (34/1497), which is similar to the literature. Verrucous carcinoma mostly arises from the glottic region and hence frequently can be diagnosed at the early stages.[7] Furthermore, regional or distant metastasis is not common. For these reasons, transoral laser surgery is often adequate with excellent survival rates.[7] In the present study, 73% of the patients had a T1a tumor, and 88% of tumors were localized in the glottic region with an overall 5-years survival rate of 100%. No neck and distant metastasis occurred in any of the patients in the study group, thus for patients with pre-operative verrucous carcinoma diagnosis, elective neck dissection is not necessary.
Laryngeal basaloid SCC is an aggressive variant of SCC which was first described in 1986.[8,9] Laryngeal basaloid SCC is often diagnosed in the supraglottic localization. Begum et al. detected the tumor in the supraglottic localization in all of the patients (8/8)[10], and Ereno et al in 86% (13/15).[11] In our patient group, the supraglottic region is the most common localization of the tumor (82%). Published studies indicated that neck metastasis occurs up to 50-70% in patients with basaloid SCC.[12,13] In our study, neck metastasis was observed in 46.4% of the patients. Due to the higher tendency to regional and distant metastasis, lower survival rates have been published. In a study published in 2008, 3 and 5-year overall survival of 15 patients reported as 50% and 38.5%, respectively.[11] In our study, 5-year disease-specific survival was determined as 69.9%. While the survival rate is 100% for the early stages, a decline to 54.5% is observed for late-stage tumors.
Spindle cell carcinoma was first described by Virchow in 1863 and in 1933 Figi described the entire pathological details.[14] The most common origin of the spindle cell carcinoma is the larynx.[15] Even though the prognosis is directly related to the stage, spindle cell carcinoma is an aggressive tumor.[14] In the present study, according to the survival rates, spindle cell carcinoma is the variant with the worst prognosis.
Laryngeal papillary SCC was first described in 1999 by Thompson et al.[16] including 104 cases as a variant of SCC with a good prognosis. Similar to the literature, survival rates are also excellent in our study. Even though, our follow-up period is inadequate to give 5-years survival rates, both of our patients are disease-free at their 17th and 28th-month follow-up. In a study in 2015, 370 patients were evaluated by using the Surveillance, Epidemiology, and End Results (SEER) Program, and the overall incidence within entire laryngeal malignancies was calculated as 0.5%.[17] In the present study, an incidence of 0.1% was calculated. High-risk Human Papilloma Virus (HPV) strains are thought to play a role in the etiology of the papillary variant of SCC.[18,19] Our lower rates can be explained by the low incidence of these strains in our country.[20] Due to the retrospective nature of our study and lack of HPV-DNA or p16 screening, we can’t make a definite interpretation.
Laryngeal lymphoepithelial carcinoma is a variant of SCC that has a resemblance to undifferentiated non-keratinized (WHO type 3) nasopharyngeal carcinoma and a high affinity to neck metastasis.[5,21] Similar to the literature, neck metastasis was observed in 75% of our patient group. According to SEER data in which 16 patients were evaluated, overall and disease specific 5-years survival of this variant was 58,8% and 63%, respectively.[22] The overall and disease specific 5-year survival in our study was assessed as 50% and 75%, which was slightly better compared to this study. However, due to our low number of patients (4), a comparison with the literature would not be suitable.
Acantholytic SCC was classified as a vSCC until 2017, however, after 2017 this variant was included in the conventional laryngeal SCC group.[5,23] The evaluation of the patients in this study group was done with the previous classification system, thus 17 patients were diagnosed with the acantholytic variant. Overall 5-years survival was calculated as 72.8% and neck metastasis was observed in 31.3% of the patients who were treated with primary surgical methods. Our results show similarity to prognosis and neck metastasis rates of the classical laryngeal SCC.[24,25]
Laryngeal adenosquamous carcinoma is a vSCC with both glandular and epithelial differentiation. Previous studies of this variant show neck metastasis up to 75% and distant metastasis up to 25% of the cases.[5] Adenosquamous carcinoma is the vSCC with one of the worst prognosis, and previous 5-years survival studies show around 30-50% survival rates.[5] Controversially, in our study, the 5-years survival rate for adenosquamous carcinoma is 83.3%. The explanation for this great difference can be explained by the absence of any patients with distant metastases at the time of diagnosis in this pathological subgroup, which will also be mentioned in the limitation of the study. Due to the absence of this feature, which is a bad prognostic factor, better survival rates were calculated.
We found that some of the patients included in this study had a preoperative pathological diagnosis, not vSCC, but conventional SCC with the incidence of 38.5-100% according to the histological subtype. It is possible to detect pathological subgroup changes in the operated patients, but it is not possible to detect this situation in the patient group who underwent non-surgical treatment. For this reason, several patients who should be in the vSCC group are considered as classic SCC, and treatment and follow-up planning can be made with this information. Another result of this condition is the survival data of our study; the survival data of some pathological subgroups were observed better than the literature. One of the possible causes of this condition may be that the patients with poor prognostic indicators, inoperable tumors and distant metastases receive non-surgical treatments, and the diagnosis of these patients should be reported as conventional SCC, not vSCC. Thus, the patient group in our study may have consisted of patients with a relatively good prognosis.
The difference between the diagnostic and definite pathology results can be explained by the heterogeneity of the tumors, which creates areas that do not contain the vSCC. Although smaller specimens are adequate for the malignity diagnosis, sub-typing is not always possible. A small biopsy is usually taken for diagnosis of laryngeal cancer. Because large biopsy fragments can cause bleeding, edema, and associated airway obstruction. In patients with small lesions planned to be treated with primary RT, permanent vocal cord damage and permanent quality of voice may occur after treatment. It can be concluded from this study that small punch biopsies may not be sufficient for the diagnosis of vSCC.
While evaluating the patients for adjuvant therapies, along with the TNM stage of the tumor, pathological subgroup change possibility should also be taken into account. Especially, the worse prognosis and high metastatic potentials of spindle cell carcinoma and basaloid type SCC should be kept in mind. Patients who are diagnosed with these two variants should be followed-up more carefully, and the physicians should be more skeptic if any recurrence suspicion is present.
The first limitation of our study is the retrospective design. The clinical and pathological staging of the tumors was reported by using the 7th Edition of The American Joint Committee on Cancer (AJCC), which was renovated by the 8th Edition in January 2018. Secondly, the vast majority of the patients who were included in the study are patients who were surgically treated. Thus, some of the patients with distant metastasis at the time of the diagnosis might be missed out. For this reason, the survival rates might be lower than our calculated results. Even though prospective studies can be designed for better prognosis and survival rate analysis, differences between pre-operative and definite pathology results will always cause under-diagnosis of variants.