Routine Computed Tomography Scanning for Tumor Staging in Advanced Laryngeal Cancer: Implications for Treatment Selection

Abstract

Determining the precise clinical extent of cancer spread within the larynx has been the most critical factor guiding treatment decisions in patients with localized laryngeal cancer. Histopathologic studies using whole organ sections have documented patterns of cancer spread and barriers to tumor growth within the larynx that provide the anatomic basis for tumor staging and function-sparing conservation laryngeal surgery. Tomographic imaging, particularly computed tomography (CT), has been integrated into clinical staging to supplement clinical observations of tumor extent and vocal cord function in determining tumor stage. Clinical tumor stage has been used routinely to make decisions for either primary surgery or organsparing radiation as definitive treatment. Traditionally, patients with early-stage (ie, stages I and II) cancers have been treated with primary radiation, and patients with more advanced cancers (ie, stages III and IV) have been treated with total laryngectomy, with or without adjuvant postoperative radiation. However, patterns of care for patients with larynx cancer continue to change. The success of partial laryngeal surgical procedures and endoscopic laser-assisted techniques has led to a significant increase in function-sparing, partial laryngeal resections for patients with early or even moderately advanced cancers. In addition, with the introduction of combined chemoradiotherapy as an alternative to total laryngectomy, organpreserving approaches have been widely adopted for patients with stages III and IV disease. Such chemoradiotherapy approaches have demonstrated successful larynx preservation that have cure rates comparable to primary surgery in the majority of patients. In patients with advanced (ie, T4) cancers, however, lower rates of organ preservation and higher complication rates have been reported, prompting many oncologists to recommend primary laryngectomy, particularly when gross cartilage invasion by cancer is evident. Thus, cartilage invasion has become an important issue in cancer staging and in treatment selection between larynx preservation and total laryngectomy. According to the sixth edition of American Joint Committee on Cancer staging criteria, minor thyroid cartilage invasion (ie, inner cortex) is sufficient to classify a tumor as T3, and invasion through the cartilage or into the extralaryngeal soft tissues would classify the tumor as T4. This has not changed in the 2010 (ie, 7th) edition, except that stage IV cancers have been divided into IVa and IVb; IVb reflects “very advanced local/regional disease” characterized by involvement of the prevertebral space, carotid artery, or mediastinal structures. The article “Prognostic Accuracy of Computed Tomography Signs for Laryngeal Cancer Patients Undergoing Laryngectomy” by Beitler et al offers a careful retrospective clinicopathologic analysis of the accuracy of pretreatment CT scanning to predict cartilage invasion and extralaryngeal cancer spread. Routine interpretation of preoperative CT scans and histopathology reports from 107 patients undergoing total laryngectomy during a 10-year period were analyzed for presence of cartilage invasion (at inner thyroid cortex) or penetration (at both inner and outer cortex). Correlations of CT signs and histopathologic findings were determined. The study confirmed that a high rate (63%) of occurrences had CT evidence of cartilage invasion or penetration, and 80% had evidence of invasion on histopathology. When CT suggested extralaryngeal extension, there was an 81% correlation with pathology findings. Interestingly, extralaryngeal spread without cartilage penetration was found in 18 (40%) of 45 occurrences. Surgeons have long understood the prognostic significance of extralaryngeal spread through thyrohyoid or cricothyroid membrane barriers, even when cartilage invasion was absent. The authors correctly cautioned that CT scanning was not highly sensitive for extralaryngeal disease. If the critical issue, however, is invasion of cartilage, only four (7%) of 54 occurrences that were read as normal thyroid cartilage had histopathologic evidence of thyroid cartilage penetration. When penetration of cartilage was read on CT, it was confirmed by histopathology in 74%, and 87% of occurrences had at least pathologic cartilage invasion. It is likely that careful serial sectioning of the whole larynx may have demonstrated an even higher correlation. Finally, the pretreatment clinical tumor stage, tumor site, or criteria used for selecting patients for total laryngectomy were not reported, rendering it impossible to assess any bias towards cartilage invasion or the frequency of either T3 and T4 cancers in the study cohort or differences between glottic or supraglottic primaries. Thus, the usefulness of CT staging in influencing the final clinical tumor stage cannot be determined. So, how should we interpret these findings, and how should CT findings influence our selection of appropriate primary treatment modality in patients with advanced laryngeal cancer? Effective tumor staging should incorporate as much clinical, radiographic, functional, JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L S VOLUME 28 NUMBER 14 MAY 1