Pretreatment neurocognitive function (NCF) status in head and neck cancer (HNC) patients (pts) with comparison to control cohort.

Abstract

5587 Background: There is increasing evidence that NCF abnormalities may occur in cancer pts. Data on pre-treatment NCF in HNC pts are lacking. This study reports NCF in pts with newly diagnosed, curable HNC compared to controls. Methods: HNC pts underwent a 2-hour battery of NCF tests prior to radio +/-chemo(bio)therapy. Domains tested were intelligence (IQ), memory, language, attention, processing speed, executive function and manual dexterity. Test performances were transformed into Z-scores using normative data (score < -1 signified deficit). Pts also had self-reported assessments for NCF, quality of life (QOL), fatigue and affect. Data obtained were compared to non-cancer controls who underwent the same tests. Results: Eighty HNC and 30 control subjects were assessed. Objective NCF testing demonstrated that HNC and control cohorts were similar across all domains, except for IQ, with pts having higher scores (mean 0.55 vs 0.12, p=0.03). However, individual analysis showed that 39% of HNC and 43% of control subjects had abnormal Z-scores in ≥ 2 domains. Multivariable analysis of factors associated with ≥ 2 abnormal NCF domains included: low education level, significant smoking history (≥ 10 pack year), previous mild brain injury, gender, and group (pt vs control). Amongst pts, HPV -ve status and non-oropharyngeal tumors were also associated with decreased NCF. Pts reported statistically worse subjective baseline symptoms compared to controls: NCF (mean FACT-COG 33.7 vs 18.2, p=0.002), QOL (FACT H&N 33.8 vs 14.9), fatigue (FACT-F 35.1 vs 15.1), anxiety (HADS 7.0 vs 3.1) and depression (HADS 3.9 vs 1.2), p<0.01 for all five parameters. Conclusions: Objectively assessed NCF was similar between HNC pts and controls, but a proportion of participants in both cohorts have multi-domain abnormal Z-scores. Several patient demographics and disease characteristics were associated with abnormal NCFs. Subjectively, pts reported worse NCF, QOL, fatigue and affect. These data suggest that participant and disease characteristics may play a larger role in determining NCF than previously shown. Whether such characteristics impact subsequent NCF is under investigation in a longitudinal study.