Patterns of Relapse and Value of Follow up Procedures in Patients with Diffuse Large B Cell Lymphoma (DLBCL) Who Achieve a Complete Remission (CR)

Purpose: Optimal follow up schedule and choice of diagnostic modalities have not been well studied in patients with DLBCL who achieve CR after initial treatment. We analyzed patterns of relapse in patients with DLBCL who were treated with CHOP±R therapy and achieved CR in our institution between 1999 and 2007. We generally follow such patients with physical exam and blood tests including LDH every 3 months and CT scans every 3 to 6 months for the first two years. Follow up schedule varied afterwards.

Patients and methods: Thirty-eight patients experienced relapse, in whom we first determined if relapse was detected by symptoms which prompted diagnostic processes such as imaging studies and/or biopsy or by prescheduled follow up blood test or systemic CT scans. Other characteristics of these patients including time from diagnosis to relapse, time from last systemic CT scan that showed no evidence of disease to relapse, and time from relapse to death were analyzed.

Results: Twenty-seven patients (71%) presented with symptoms (“symptomatic patients”); 12 with palpable superficial masses, 5 with local pain, 5 with neurological symptoms due to central nervous system (CNS) involvement, 2 with persistent fever, 1 with gastrointestinal bleed, 1 with nasal congestion and 1 with persistent cough. Eleven patients (29%) were asymptomatic when relapse was detected (“asymptomatic patients”). Two showed elevated serum LDH levels which prompted imaging studies leading to the diagnoses. Nine were found to have relapsed diseases by prescheduled systemic CT scans. Among 27 symptomatic patients, 7 (18% of all relapses) had diseases that were not detectable by CT scans (5 isolated CNS relapses, 1 GI relapse, and 1 subcutaneous relapse in upper extremity). Patient characteristics at the time of initial diagnosis were similar in the two groups. Median time from initial diagnosis to relapse in symptomatic and asymptomatic patients were 11.9 months (range 3.8–98.1) and 12.5 months (range 7.2–50.7), respectively (p=0.635). Median time from last systemic CT scan to relapse were 2.9 months (range 0.1–33.0) and 5.2 months (range 1.9–10.8), respectively (p=0.895). Time from relapse to death of any cause were similar in the 2 groups (log-rank p=0.423) but the median time was longer in asymptomatic patients (16.6 months vs 39.8 months). Since late relapses potentially tend to be detected by symptoms due to infrequent CT scans, analyses were next limited only to those who relapsed within 2 years after diagnosis. The results were similar between symptomatic and asymptomatic group.

Conclusions: This study showed that more than two thirds of patients were symptomatic at the time of relapse. And up to 18% of relapsed diseases were undetectable by routine CT scans. Half of “symptomatic” patients noticed their symptoms less than three months after last CT scan, suggesting that even every 3 months CT would miss these relapses before patients were symptomatic. Longer median survival (though the overall difference was not significant) in “asymptomatic patient” could be attributed in part to “early detection of relapse and intervention” but in contrast, asymptomatic patients might have had less aggressive diseases allowing prescheduled CT scans to detect the diseases before they are symptomatic. Important question is whether delaying detection of relapse and treatment till patient is symptomatic would be associated with worse clinical outcome. This question is also translated into whether routine CT scan is even necessary, and if so, what is the optimal schedule of imaging studies. Answers to these questions may depend on the estimated risk of relapse based on various prognostic factors. Larger scale studies are needed in this area.