1180 - The Impact of Imaging-Detected Extranodal Extension on Treatment De-Escalation for HPV-Related Oropharyngeal Cancer: A Pooled Analysis of Prospectively Treated Patients
Presenter(s)
M. Quintana1, F. Abodunrin2, O. Pasternak-Wise3, M. Muzaffar3, I. Hussain3, A. Juloori4, D. J. Haraf4, L. Chen5, N. Choudhury6, A. Pearson6, H. Arshad7, C. Lin3, E. Izumchenko3, N. Agrawal7, E. E. Vokes6, A. J. Rosenberg8, and R. R. Katipally4; 1University of Chicago, Chicago, IL, United States, 2The University of Chicago, Chicago, IL, 3University of Chicago, Chicago, IL, 4Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL, 5Department of Radiation Oncology, University of Chicago, Chicago, IL, 6Department of Medicine, Section of Hematology/Oncology, University of Chicago, Chicago, IL, 7Section of Otolaryngology-Head and Neck Surgery, Department of Surgery, University of Chicago, Chicago, IL, 8Department of Medicine, Section of Hematology and Oncology, University of Chicago, Chciago, IL
Purpose/Objective(s): It remains poorly understood if patients with HPV-related oropharyngeal squamous cell carcinoma (HPV-OPSCC) exhibiting imaging-detected extranodal extension (iENE) are appropriate candidates for treatment de-escalation strategies. The purpose of this pooled analysis was to correlate iENE with treatment outcomes. We hypothesized that iENE was associated with low rates of locoregional failure (LRF) in the context of de-escalation.
Materials/Methods: This was a pooled analysis of patients with locally advanced HPV-OPSCC treated on response-adapted de-escalation trials: OPTIMA (NCT02258659), OPTIMA II (NCT03107182), NCT04572100, and an off-protocol prospective registry. Very low risk patients (T0-2N0-1 with single LN < 3cm) were excluded. All patients received induction chemotherapy x 3 cycles (carboplatin plus paclitaxel or nab-paclitaxel +/- nivolumab) and underwent RECIST response assessment. High Risk patients were defined as T4, N2 (contralateral LNs), N3, or tobacco smoking > 10-20 pack-years, and received 45-50 Gy chemoradiation (Intermediate Dose Arm [IDA]) to reduced volume if =50% response or 70-75 Gy chemoradiation (Regular Dose Arm [RDA]) if <50% response. All others were considered Low Risk and received 50 Gy RT alone to reduced elective volume or transoral robotic surgery (Low Dose Arm [LDA]) if =50% response, IDA if 30-50% response, or RDA if <30% response. iENE was graded on baseline MRI or CT per HNCIG consensus guidelines. LRF, regional failure (RF), and distant metastasis (DM) were analyzed using Fine-Gray models (death was a competing risk). Progression-free survival (PFS) and overall survival (OS) were analyzed using Kaplan-Meier.
Results: 234 patients were included (51% High Risk, median follow-up 4.0 years). 83% received treatment de-escalation (37% LDA, 46% IDA, 17% RDA). iENE was present in 61% (HNCIG grade 1 [25%], grade 2 [22%], grade 3 [13%]), with iENE grade 2+ in 35%. At 3 years, OS, PFS, LRF, RF, and DM were 97%, 93%, 4%, 2%, and 2%, respectively. Any iENE and grade 2+ iENE were not associated with LRF (any iENE: sHR 1.1, P = 0.87; grade 2+ iENE: sHR 2.0, P = 0.33) or RF (any iENE: sHR 1.0, P = 0.99; grade 2+ iENE: sHR 1.3, P = 0.76) [Table 1]. In grade 2+ iENE, 3y LRF was 0% in the low and regular dose arms, but 10% in the intermediate dose arm (4 events; 2 RFs; all occurring in high risk patients; all receiving induction without nivolumab [P = 0.28]). Grade 2+ iENE numerically trended toward increased DM (sHR 6.1, P = 0.11).
Conclusion: In this pooled analysis of response-adapted de-escalation for HPV-OPSCC, consisting of higher risk patients, overall outcomes were excellent. Absolute rates of LRF were low even in the presence of iENE, supporting continued inclusion of such patients in future de-escalation trials. A subset of high risk patients with grade 2+ iENE may remain vulnerable to both LRF and DM, warranting close examination.
| Grade 2+ ENE (95% CI) | Not Grade 2+ ENE (95% CI) | |
| 3y LRF | 6% (2% - 13%) | 3% (1% - 6%) |
| 3y RF | 3% (1% - 9%) | 2% (1% - 5%) |
| 3y DM | 5% (1% - 12%) | 1% (0% - 4%) |