← Back to List

Program Number: B307
Session Name: Poster Session

The Influence of Perineural Invasion on Local Control From Adjuvant Radiation Therapy Among Low- And Intermediate-grade Major Salivary Cancers

Alexandra T Bourdillon, MD¹; Peter Callas, BA²; Mirabelle Sajisevi, MD²; Karolina Plonowska-Hirschfeld, MD³; William R Ryan, MD¹; ¹University of California - San Francisco, Department of Otolaryngology - Head & Neck Surgery; ²University of Vermont, Division of Otolaryngology, Department of Surgery, The Larner College of Medicine; ³Washington University School of Medicine at St. Louis, Department of Otolaryngology

The utility of adjuvant radiation therapy (aRT) in early-stage low- and intermediate grade salivary gland carcinomas with perineural invasion (PNI) in the absence of other high-risk features has not been well studied, likely given the rarity of this disease subtype. 

We conducted a secondary analysis of a multi-institutional retrospective cohort study organized by the American Head and Neck Society Salivary Gland Section of 336 patients with T1-2 low- and intermediate-grade major salivary gland carcinomas from 2010 to 2019 across 40 international centers, excluding patients with intraoperative tumor spillage, positive margins, lymph node metastases, distant metastases, lymphovascular invasion (LVI) or unreported PNI/LVI (see Figure 1).  

Isolated PNI as the only high-risk feature was observed in 24 (7%) patients including intratumoral (6/24 or 25%), extratumoral (2/25, 8%), or not otherwise specified (NOS) PNI (16/25, 67%).  On univariate testing (Fisher’s exact test), the isolated PNI and control patient cohorts did not significantly differ between sex, age group, salivary gland subsite, T classification, histopathologic grade, facial nerve sacrifice, and local recurrence rate (see Table 1).  Patients with isolated PNI were more likely to undergo aRT (42% versus 18%, p =0.01).  Isolated PNI was not significantly associated with higher local recurrence rates (2/24, 8%) compared to the control cohort (6/312, 2%) (Hazard Ratio (HR): 2.94, 95% CI: 0.59-14.64, p=0.19). For the isolated PNI cohort, one (out of 10, 10%) who underwent aRT experienced a local recurrence and one (out of 14, 7%) who underwent observation experienced a local recurrence.  The median follow-up duration for the aRT and observation groups was 55 months (IQR: 38-70 months), and 56 months (IQR: 8-66 months), respectively. Both patients who had local recurrence had PNI NOS. For isolated PNI, aRT was not significantly associated with local recurrence reduction (HR: 1.23, 95% CI: 0.08-19.86, p =0.89).  Multivariate analysis of the study cohort (n=336) showed no significant association of aRT with reduction of local recurrence after controlling for PNI (HR: 0.39, 95% CI: 0.05-3.29, p=0.38), and after controlling for facial nerve sacrifice (HR: 0.34, 95CI: 0.04-3.06, p =0.33).

This study suggests that aRT does not necessarily reduce local recurrence in early-stage low- and intermediate grade salivary gland carcinomas with PNI in the absence of positive margins, LVI, and lymph node metastases.  Further prospective and randomized studies are needed to evaluate the oncologic safety of post-surgical observation of early-stage low- and intermediate grade salivary gland carcinomas with isolated PNI.  We acknowledge that this study may be affected by potential selection bias given the retrospective nature and limited by sample size, even in a multi-institutional cohort of this magnitude.

Figure 1. Study Cohort Flow Chart

 

Table 1. Cohort Overview

 

 

← Back to List