[1220] Validation of the Risk Model in a New Patient Cohort with HNSCC
M Brandwein-Gensler, B Wang, C Penner, R Smith, M Prystowsky, N Schlecht, A Negassa. University Alabama, Birmingham, Al; NYU, NY, NY; University Manitoba, Winnipeg, MB, Canada; Montefiore/Einstein, Bronx, NY; Einstein, Bronx, NY
Background: Some patients with low-stage (LS) HNSCC are expected to fail single modality therapy, indicating that more aggressive treatment would have been warranted. Identifying LS patients at risk for disease progression (DP), for multimodality treatment, would represent an important advance. Our Risk Model may aid in developing new treatment paradigms. Here we report the performance of our model in a new multi-center cohort.
Design: Eligible patients were identified and resection slides were reviewed (MBG). Kaplan-Meier (KM) analysis was performed for DP and OS, stratified by risk category. Stratified Cox proportional hazard regression was performed, adjusted for potential confounders.
Results: Table 1 summarizes the cohort.
| MMC | NYU | Manitoba | |
| Number SCCs | 225 | 30 | 56 |
| Oral (%) | 48 | 93 | 93 |
| Oropharynx (%) | 24 | 0 | 7 |
| Larynx(%) | 28 | 7 | 0 |
| Age ீ60 (%) | 37 | 47 | 45 |
| Low stage (%) | 37 | 50 | 46 |
| High stage (%) | 63 | 50 | 54 |

| HR | 95% CI | p | |
| High-risk | 2.20 | (1.14, 4.25) | 0.019 |
| Female | 0.80 | (0.46, 1.41) | 0.442 |
| Age ு 60 | 1.31 | (0.77, 2.26) | 0.323 |
| Margin ∠ 5 mm | 1.43 | (0.78, 2.62) | 0.242 |
| Oropharynx | 0.71 | (0.31, 1.64) | 0.425 |
| Larynx | 0.76 | (0.35, 1.66) | 0.490 |
| T = 3/4 | 1.64 | (0.89, 3.02) | 0.112 |
| LN+ | 1.01 | (0.55, 1.87) | 0.963 |
| No adjuvant RT | 1.37 | (0.74, 2.54) | 0.322 |