xRead - Nasal Obstruction (September 2024) Full Articles

20426984, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/alr.23262, Wiley Online Library on [02/01/2024]. 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

196

KUANetal.

treatment. The unique characteristics of IP are discussed previously in this consensus statement, and the focus of this section is on malignant transformation of IP. The over all incidence of identifying IP-transformed SCC (IP-SCC) accounts for 2%–10% of all IP cases, with synchronous lesions much more likely to develop than metachronous ones. 247,653,1107–1109 There is evidence to suggest that IP-SCC carries a more favorable prognosis compared to de novo SCC (DN-SCC, or primary sinonasal SCC not arising from pre existing IP). 1110–1112,1121 A recent meta-analysis has found that patients with DN-SCC carry a 1.87-fold increased risk of mortality with 5-year OS of 56%, compared to 65% for patients with IP-SCC. 1113 Despite the improved reported survival, IP-SCC patients often present with locally advanced tumors (74% with T3/T4 stage disease), and 23.8% cases experienced a recurrence despite simi lar treatment to DN-SCC. 1114 It should be noted that most studies are limited by sample size, heterogeneity in tumor origin, clinical stage, and treatment modalities. Further more, the significance of the findings is most vulnerable to the studies’ respective methodology in classifying the subjects into their respective cohorts. IP, and by exten sion, IP-SCC, is defined histologically. To date, there is no defined set of molecular markers to distinguish IP-SCC from DN-SCC. Accurate classification of IP-SCC requires presence of benign IP in cases with carcinoma (syn chronous lesions) or history of IP by report (metachronous lesions). Efforts to further characterize these tumors by molecular pathways should take precedence, with large multicenter studies required to validate historic findings (Table XXI.B.1). Aggregate grade of evidence : C (Level 4: eight studies) Recently, noninvasive techniques, such as preoperative radiographic imaging, have been studied that better pre dict the presence of IP-SCC. Focal hyperostosis on CT scans has been correlated with tumor origin and site of attachment, and bony erosion is indicative of aggressive tumor. 711,1120 In addition, unique characteristics identified in MRI have been used to better differentiate IP from IP-SCC. Benign IPs demonstrate a distinct morphologic pattern on MRI known as convoluted cerebriform pat tern (CCP), which is appreciated as alternating hypo and hyperintense bands on T2-weighted and contrast enhanced T1-weighted imaging. 709,1122 Currently, a few studies suggest that CCP may be beneficial in distinguish ing IP from IP-SCC. 1123–1126 Specifically, there appears to be an overall loss of CCP among IP-SCC, with most tumors 2 Imaging to predict malignant transformation

demonstrating either partial or near-complete absence (Table XXI.B.2). The obvious limitation in making diag noses based on CCP is that it is subject to interpretation error. A more objective analysis utilizes ADC value, and it has been found that malignant sinonasal neoplasms have lower mean ADC values. 1127,1128 ADC map generation from diffusion-weighted MRI (DW-MRI) images requires additional image analysis software, but current evidence supports its potential role in developing a predictive model for IP-SCC. Aggregate grade of evidence : C (Level 4: six studies) IP shows an inverted (endophytic), nondestructive growth into the underlying stroma, enclosed by an intact basement membrane. Benign areas may be concurrently present with zones of malignant tumor, showing considerable volume variation between cases. IP is cytologically bland, shows normal maturation and polarization, may have koilocytic atypia, and contains intraepithelial mucous cysts filled with mucin and cellular debris, with prominent transmi grating intraepithelial neutrophils. Areas of dysplasia may develop as a precursor to malignant transformation. 653,1129 Hyperkeratosis, increased basaloid morphology, conspicu ous pleomorphism, and increased mitoses, including atyp ical forms, with loss of transepithelial neutrophils signal malignant transformation. Carcinomas include nonkera tinizing, keratinizing, and low-grade papillary subtypes, and even SNUCs. 653,664,666,1130–1134 Keratinizing carcinoma may appear as verrucous carcinoma, although most ker atinizing carcinoma are represented as well, moderately, or poorly differentiated. Recognized as a gradient with out any absolute morphologic cutoffs, it is important to stress the volume of the neoplastic tissue (i.e., signifi cantly more epithelium than stroma) and the pleomor phism, destructive growth, and atypical mitoses that define carcinoma, without any one feature alone being suffi cient. Other malignant features include destructive bone invasion, paradoxical basal maturation, and true tumor comedonecrosis. 648,653,666,667,1135 True invasion is seldom identified, but when there is a desmoplastic reaction, loss of basement membrane, single cell infiltration, irreg ular, stellate islands of neoplastic cells, and destructive bone/cartilage invasion, definitive traditional invasion can be confirmed. p53 may be abnormally overexpressed, but a completely negative reaction can also be seen. 655,1136 There is no Ki-67 proliferation index cutoff for diagnos ing carcinoma. 653 p16 may be positive, but few cases show transcriptionally active high-risk HPV by in situ RNA hybridization. 653,655,657,666,667,1137 3 Histopathology and molecular and genetic studies