In:
Journal of Clinical Oncology, American Society of Clinical Oncology (ASCO), Vol. 31, No. 15_suppl ( 2013-05-20), p. 3635-3635
Abstract:
3635 Background: RECIST is widely used to evaluate anticancer therapy efficacy, yet its response cutoff values have not been proven biologically or measurement-error relevant. Our previous study showed that the variability in measuring relative change in total tumor burden (TTB%) was +/-10% in metastatic colorectal cancer (mCRC). Our study investigated the impact of a finer gradation of response categories in predicting survival. Methods: 468 patients enrolled in a phase II/III clinical trial evaluating a systemic therapy in mCRC were analyzed. TTB on baseline and 6-week (+/- 3-wk) scans were obtained per RECIST 1.0. Overall survival (OS) was defined from the start of treatment and the date of the scan using the landmark method. The TTB% was summarized using RECIST category and a finer gradation that included cut-offs established in our variability study ( 〈 -30%, -30% - -10%, -10% - 10%, 10% - 20%, 〉 20%). The OS and TTB% correlation was evaluated by the Kaplan Meier method and Cox regression. The discriminatory powers of the response summaries were examined using Harrel’s c statistics and concordance probability. Results: Out of the 468 patients, 141 died. The median survival time for patients with a TTB% at 6 weeks of [-30%, -10%] (n=116), [-10%, 10%] (n=171), [10%, 20%] (n=43), 〉 20% (n=62) were 417, 287, 223 and 172 days, respectively. Among those with a change of 〈 -30% (n=76), over half of patients were still alive. The hazard ratio for OS compared to those with a change of 〈 -30% are listed in the Table. The inclusion of additional categories increased both Harrel’s c-statistic (0.69 vs 0.73) and concordance probability (0.63 vs 0.69). Combining the [-10%, 10%] and the [10%, 20%] categories yielded very similar statistics compared to having all 5 categories. Both definitions of OS yielded consistent results. Conclusions: Evidence suggests that, in the context of this study, the minor change category of [-30%, -10%] at 6-wk correlates with longer survival compared to the change category of [-10%, 20%] , suggesting a possible re-evaluation of conventional response cut-off values. [Table: see text]
Type of Medium:
Online Resource
ISSN:
0732-183X
,
1527-7755
DOI:
10.1200/jco.2013.31.15_suppl.3635
Language:
English
Publisher:
American Society of Clinical Oncology (ASCO)
Publication Date:
2013
detail.hit.zdb_id:
2005181-5
