Prospective Trial of Personalized Fractionation in Low-risk HPV Positive Oropharyngeal Cancerroph
OC-0106
Abstract
Prospective Trial of Personalized Fractionation in Low-risk HPV Positive Oropharyngeal Cancerroph
Authors: Jimmy Caudell1, Michelle Echevarria1, George Yang1, Youngchul Kim2, Kedar Kirtane3, Julie Kish4, Jameel Muzaffar3, Christine Chung3, Heiko Enderling5
1Moffitt Cancer Center, Radiation Oncology, Tampa, USA; 2Moffitt Cancer Center, Biostatistics, Tampa, USA; 3Moffitt Cancer Center, Head and Neck/Endocrine, Tampa, USA; 4Moffitt Cancer Center, Personalized Medicine, Tampa, USA; 5Moffitt Cancer Center, Mathematical Oncology, Tampa, USA
Show Affiliations
Hide Affiliations
Purpose or Objective
An early imaging response during radiotherapy for head and neck cancer is associated with progression free survival (PFS). We developed a mathematical model of pre-treatment growth and radiation response dynamics, with the hypothesis that the model output, proliferation saturation index (PSI), could select the most favorable fractionation [hyperfractionation (HFx) or conventional (QD)] to elicit a rapid imaging response at four weeks during treatment.
Material and Methods
Patients with T0-2 N0-1 M0 p16-positive low-risk OPC were enrolled (NCT03656133). PSI was calculated from pre-treatment diagnostic and simulation CT scans, with patients > 0.75 assigned to HFx. This protocol was a single arm, phase II design with a primary endpoint of improving ≥ 32% reduction in tumor volume by week 4 to 63%, compared to a historical control of 49%. Secondary endpoints included acute toxicity per CTCAE v5, imaging response at 2-3 months post treatment, and PFS and overall survival (OS).
Results
From 10/2018 – 2/2022, 55 patients were enrolled with select pretreatment characteristics listed in Table 1. Based on PSI, HFx was assigned in 42 patients (76.4%) and QD in 13 patients (23.6%). Response at the target lesion at 4 weeks was a median 40.1% reduction (range 100% reduction – 80.4% growth). 32 of 55 (61.2%) patients met the criteria ≥ 32% reduction in tumor volume by week 4, meeting criteria for efficacy. Only 5 patients (9.1%) developed acute CTCAE v5 grade 3 toxicity attributable to RT; there were no grade 4-5 adverse events (AEs). All grade 3 acute AEs had resolved at last follow-up, with no late grade 3-4 AEs reported. Of the 53 evaluable patients at 2-3 months post-treatment, 52 (98.1%) had a complete anatomic and/or metabolic response, and one (2.1%) had a partial anatomic/metabolic response. Median follow-up was 20 months at time of analysis. For the entire cohort, actuarial PFS at 2 years was 91.3%. Patients with a rapid imaging response had 2-year PFS 91.6%, compared with slow responders of 90.9% (p=0.22).
Patient Characteristics | N (%)
|
Median Age
| 64 (range 30-82)
|
Sex Male Female |
45 (81.8) 10 (18.2) |
ECOG PS 0 1 |
46 (83.7) 9 (16.3) |
Primary Tumor Site Base of Tongue Tonsil Soft Palate Unknown Primary |
31 (56.4) 22 (40) 1 (1.8) 1 (1.8) |
T Category 0 1 2 |
1 (1.8) 31 (56.4) 23 (41.8) |
N Category 0 1 |
6 (10.9) 49 (89.1) |
| Prior Malignancies | 14 (25.5) |
Smoking History Never Former/Current |
30 (54.5) 25 (45.5) |
Conclusion
Selection of personalized radiotherapy fractionation using the PSI model appears to be a promising approach, with an improvement in the percentage of patients achieving a mid-treatment imaging response compared to historical controls. Grade 3 acute toxicity was seen in 9.1% of patients, with no patients requiring a gastrostomy tube related to therapy. Initial PFS results with this approach appear favorable.