Main Session
Sep 29
PQA 07 - Head and Neck Cancer, Lung Cancer/Thoracic Malignancies, and Nursing and Supportive Care

3436 - Clinical Predictors of Reactive Gastrostomy Tube Placement During Radiation Therapy for Locally Advanced Head and Neck Cancer

03:45pm - 05:00pm ET
Poster Hall - Exhibit Hall A
Screen: 2
POSTER

Presenter(s)

Paul Buclez, MS, BS - California Northstate University College of Medicine, Elk Grove, CA

P. Buclez1, A. Kallesh2, J. D. Tucker3, L. Chapman3, S. K. J. Kim4, L. Evangelista3, and S. S. D. Rao4; 1California Northstate University College of Medicine, Elk Grove, CA, 2Idaho College of Osteopathic Medicine, Meridian, ID, 3University of California Davis, Sacramento, CA, 4University of California Davis, Department of Radiation Oncology, Sacramento, CA

Purpose/Objective(s):

Reactive gastrostomy tube (G-tube) placement is a common supportive intervention during radiation therapy (RT) for head and neck cancer (HNC), yet predictors of reactive placement remain incompletely defined. We evaluated baseline clinical and treatment-related factors associated with G-tube placement during RT or within three months of treatment completion.

Materials/Methods:

In this IRB-approved retrospective study, we reviewed patients with locally advanced HNC treated with definitive or adjuvant RT between July 2014 and July 2023. Reactive G-tube placement was defined as insertion during RT or within three months of RT completion. Patients with pre-treatment G-tubes or placement beyond three months were excluded. Baseline demographic, functional, treatment, and social factors were analyzed using univariate logistic regression and Chi Square. Variables of interest were included in multivariate logistic regression. Statistical significance was defined as p<0.05.

Results:

Of 227 patients identified, 210 met inclusion criteria. The cohort included 152 males (72.4%), with a median age of 63.5 years (range 17–94). AJCC stage III–IV disease was present in 144 patients (68.6%). Concurrent chemotherapy was administered to 137 patients (65.2%), and 97 (46.2%) underwent surgical resection. Median Karnofsky Performance Status (KPS) at consultation was 90 (range 50-100). Baseline dysphagia was documented in 108 patients (62.8%), with dysphagia status available for 172 patients. Reactive G-tube placement occurred in 35 patients (16.7%).

On univariate analysis, lower KPS (OR 0.57 per 10-point increase, p=0.003), higher PAS score (OR 1.28 per point increase, p=0.004), baseline dysphagia (OR 3.07, p=0.02), and concurrent chemotherapy (OR 3.00, p=0.02) were associated with reactive G-tube placement. AJCC staging (p=0.43), smoking (p=0.57) and surgery (p=0.055) did not correlate.

On multivariate analysis, lower functional status remained independently predictive: for every 10-point increase in KPS, the odds of reactive G-tube placement decreased by 39% (OR 0.614, p=0.035). Chemotherapy was also independently associated with increased odds of G-tube placement (OR 3.28, p=0.03).

Conclusion:

Lower baseline functional status and chemotherapy independently predicted reactive G-tube placement during or shortly after RT for locally advanced HNC. These findings underscore the importance of early functional assessment and may help identify patients who may benefit from earlier G-tube placement prior to treatment initiation and reduce treatment-associated complications. Ongoing studies will seek to expand the cohort of patients we analyze and develop a more robust predictive model to aid in patient care.