Unanswered Questions on the Optimal Extent of Tongue Cancer Resections

Abstract

Owing to the site- and stage-dependent molecular changes beyond the excised surgical margins of mucosal head and neck squamous cell cancers [HNSCC], an absolute cutoff for safe margins is difficult to define. Entrapment of the primary tumor in a specified compartment by a barrier clearance concept can circumvent this to a considerable extent, but it is not possible in all sites. A case of recurrent squamous cell cancer (SCC) of the tongue which had undergone wide excision of the lesion twice and later required a total laryngectomy because of crossover of the recurrent disease to the preepiglottic space and thereby to the glottis-supraglottic region is presented as an example to illustrate this predicament.

Repeated surgery for multiple metachronus primary carcinomas of the tongue is a daunting task and often presents a difficult therapeutic dilemma to the head and neck disease management team. A 51-year-old ex-smoker with Eastern Cooperative Oncology Group (ECOG) performance status (PS) 2 presented to us with moderately differentiated squamous cell carcinoma (MDSCC) of the left lateral border of the tongue, staged as T2 N1 M0 on 8/9/2011. On examination, there was an ulceroproliferative growth of size 3 × 3 cm on the left lateral border, 4 cm from the tip of the tongue. The floor of mouth was free and the posterior limit of the ulcer was palpable. A 2 × 2-cm-sized lymph node was palpable at level 2. Wide excision of the lesion was done with an extended supraomohyoid neck dissection. The final pathological TNM was pT2N2bM0, the tumor had a depth of infiltration (DOI) of 1.8 cm, the tumor margins were free, and level 3 and 5 nodes were positive. Post-operative radiotherapy was given to both sides of the neck to the dose of 60 Gy with an electron boost to the lower neck. He was disease-free till 16/10/2018 when he presented with a lesion at the right lateral border of the tongue of size 3 × 2.5 cm extending from the tip to the base of tongue with palpable posterior limit and minimal encroachment to the floor of the mouth, crossing the midline with no nodes, which was biopsy-proven to be MDSCC.ECOG was 1 and after a delay of 1 month due to the indecision of the patient, we did a wide excision of the lesion and reconstruction with a nasolabial island flap [1] harvested from right side (Fig. 1), and the final HPR showed DOI of 0.8 cm with adequate tumor margins. After 8 months of close follow-up, he was found to have an ulceroproliferative growth involving both lingual and laryngeal surfaces of the epiglottis and aryepiglottic folds on both sides with slough in the vallecula and the disease was designated and staged as supraglottic cancer, T3N0M0. In an MDT setting with an informed consent of guarded prognosis, a decision was taken to perform an extended total laryngectomy (TL), base-of-tongue (BOT) resection after a careful contrast-enhanced CT scan study. The final salvage procedure amounted to a TL with BOT resection partial pharyngectomy and reconstruction with a PMMC flap (patch) and the HPR was PT4 due to thyroid cartilage invasion by the tumor (Figs. 2a and b). With an uneventful post-operative period and restored swallowing function, he is now on follow-up.

Fig. 1.

Late appearance of wide excision of the recurrent lesion and reconstruction using a nasolabial island flap. Almost the whole of the tongue is replaced by the transforming (mucosalized) skin

Fig. 2.

a. Second recurrence in the pre-epiglottic space extending to the glottic-supraglottic region necessitating total laryngectomy with the tumor resection. b. The defect of the laryngectomy and excised remnant base of tongue resurfaced with a pectoralis major myocutaneous flap

In tongue cancer resections, even when surgical resected margins are histologically normal, it may harbor molecular signatures of cancer fields suggesting that some tissue progresses towards malignancy perhaps in the presence of a favorable microenvironment. As per English literature, 52% of patients with histopathological negative margins had p53 gene mutation at the margin and 38% of those cases with positive margins after molecular analysis had local recurrence whereas the negative cases did not recur [2]. Our own complementary evaluation studies of biopsy have found nuclear and cytoplasmic expression of NF-kappaB and COX-2 at the surgically negative margins [3]. Complementary molecular profiling of margins may reduce recurrences but not the development of second field tumor (SFT) or second primary tumor (SPT) for which most recently tumor immunity [4] or microenvironment/biomic [5] manipulation strategy is being tried. Some of our other relevant works on molecular epidemiology which has a bearing on tumor relapse includes SNP (single nucleotide polymorphism), cyclin D1 gene, Rb (retinoblastoma) pathway (genomic study), CDK6 (cyclin-dependent kinase), p21, P27, TGFb (tumor growth factor) protein, LOH (loss of heterozygosity), DGS162 [as predictors of disease-free survival (DFS)], micro alterations in qp21 GX2, and loss of CDKN2A and CDKN2B expression (as predictors of recurrence) [6–9]. We are also on the lookout for answers to some of our intriguing clinical observations like absence of lymph node metastasis in spite of local tumor progression and aggressive lymph nodes in early lesions, at a molecular level with some breakthrough which concurs with observations worldwide. Wolf et al. [10] correlated invasion to the presence of CD4, CD8, and Fox P3 positive T cells. Likewise, the disruption of E-cadherin function and increased expression of vimentin and the transcriptional oncogene, SOX2 was causally related to epithelial mesenchymal transformation (EMT) tumor progression and metastasis by Lamouille et al. [11].

For the very same reasons discussed above, the optimal margins of resection and the role of frozen section control have remained controversial [12]. Perhaps, an adequate clearance with careful consideration of tumor biology, stage, and treatment setting (primary/salvage) is more important than the actual achievement of a magic figure of .5 cm all around the tumor. Nevertheless, our experience with pathological target of three-dimensional metric margins of .5 cm with the help of intraoperative frozen section, especially in repeat surgeries where we work with a limited scope for wider clearance has been encouraging as in the case described here. We have been using frozen section controls for more than a decade and a cross-over study of its utility is on the pipeline. Furthermore, in situations where the tumor is distinctly invasive with the front of the tumor likely to be close to the resected margin, we would resort to a compartmental resection and reconstruction with skin flap [1, 13].

Funding Information

This study was supported by the Kerala State Council for Science Technology and Environment.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.