Comparing and validating simple measures of patient-reported peripheral neuropathy (PRPN) for NCCTG Clinical Trials: A pooled analysis of 2,440 patients (pts)

Abstract

9534 Background: The introduction of new anti-cancer therapies which cause pts to incur PN has required the development of new methods to assess this treatment outcome. This study compared and validated alternative methods of assessing PRPN in a series of North Central Cancer Treatment Group (NCCTG) clinical trials. Methods: Data were combined from 5 NCCTG trials (2440 pts) that used single- item questions scaled 0–100 relating to numbness and tingling respectively in fingers and toes to measure PN. Patient quality of life (QOL) was assessed using UNISCALE, Symptom Distress Scale (SDS), Profile of Mood States (POMS), Brief Pain Inventory (BPI) and Subject Global Impression of Change (SGIC). Associations between items were tested using Bland-Altman method and correlation. Wilcoxon tests compared QOL scores between pts with PN (score >40) vs. no PN. Change in PN and QOL was compared with a 20-point change defined as clinically meaningful. Minimal important differences (MID) were estimated using both anchor-based approach and distribution-based methods. Results: At baseline, the proportion of PRPN was 11% (numbness) and 10% (tingling); and 18% on each at last assessment. The correlation between PN items at baseline and last assessment was 0.8, with an average difference of <1 point on the 100-point scale. The MID estimate of SGIC in overall QOL, Physical Condition and Emotional State were 24, 21, 25 for numbness (respectively); and 19, 18, 17 for tingling (respectively). The distribution-based MID estimate was 12 for both PN items. Pts with substantial PN reported an average of 10 points lower QOL, mood and worse SDS and a 20 point-deficit in the BPI interference items. Pts with ≤20 points worsening in PN reported significant declines in SDS (p<0.001), worst pain (p<0.02), but not in POMS (p=0.64) or overall QOL (p=0.86). Conclusions: The two PN items for numbness and tingling were redundant. Anchor-based MID methods produced larger and inconsistent estimates relative to the distribution-based methods. Evidence of criterion validity and responsiveness indicates that these simple measures of PN can be used successfully in cancer clinical trials. No significant financial relationships to disclose.