Background

The use of routine surveillance imaging (RSI) for patients in first complete remission (CR1) following front-line rituximab (R) based anthracycline therapy remains controversial. We compared patients with diffuse large B-cell lymphoma (DLBCL) who received an R-CHOP or a similar regimen, obtained a CR and then were followed by either RSI or clinical surveillance (CS) in which scans were only performed for signs or symptoms.

Methods

Patients from three tertiary care center from 2001-2011, who achieved a CR1 with frontline R-CHOP or similar therapy for DLBCL, and had a minimum follow up of 1 year were analyzed. Patients with HIV-related lymphoma, transformed lymphoma, and post-transplant lymphoproliferative disorders were excluded. Patients with composite lymphoma were included only if the DLBCL component was >50%. Patients were stratified into two groups based on the surveillance strategy employed. Baseline patient characteristics, prognostic features, treatment type, and outcomes were compared.

Results

391 patients with DLBCL treated with R-CHOP or similar regimens who obtained CR1 were analyzed. There were 129 patients in the CS group and 262 in the RSI group. Patient characteristics (age, gender, stage, and IPI) were similar in the two groups. The median follow up is 5 years (range 1 – 12). Relapse after CR1 was detected in 26 (20%) of patients in the CS group and 46 (18%) of the RSI group. The median number of images in the CS group was 0 (range 0-14) and 4 (range 1-27) in the RSI group, p<0.0001. The median average number of images per year of follow up in the CS group was 0 (range 0-6) and 1 (range 1-13) in the RSI group, p<0.0001. Relapses were detected through clinical manifestations in 100 % of CS cases versus 43% in RSI cases, p=0.01. The 5 year progression-free survival (PFS) was 76% in the CS group and 82 % in the RSI group (p = 0.31). The 5 year overall survival (OS) was 87% in the CS group and 92 % in the RSI group (p=0.15). The table shows an analysis of OS by IPI and type of surveillance.

5 year Overall SurvivalCS GroupRSI Groupp-value
Low IPI (n=174) 94 (84-98) 97 (90-99) 0.53 
Low-Int IPI (n=86) 84 (57-95) 88 (74-94) 0.71 
High-Int IPI (n=93) 78 (29-93) 90 (68-97) 0.25 
High IPI (n=38) 74 (29-93) 78 (53-91) 0.83 
5 year Overall SurvivalCS GroupRSI Groupp-value
Low IPI (n=174) 94 (84-98) 97 (90-99) 0.53 
Low-Int IPI (n=86) 84 (57-95) 88 (74-94) 0.71 
High-Int IPI (n=93) 78 (29-93) 90 (68-97) 0.25 
High IPI (n=38) 74 (29-93) 78 (53-91) 0.83 
Conclusions

The majority of relapses in patients with DLBCL after achieving CR1 to an R-CHOP or similar regimen occur when signs or symptoms of the disease lead to evaluation and are not detected by RSI. Although asymptomatic relapses are occasionally detected by RSI, in this large cohort of patients neither a PFS nor OS benefit could be demonstrated in favor of RSI. Given the additional cost, radiation exposure and risk of additional procedures, we conclude that the use of RSI in patients with DLBCL in CR1 has limited clinical utility.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution