Active Surveillance Strategies Have Neither Clinical or Survival Benefit Following High-Dose Chemotherapy and Progenitor Cell Rescue in Follicular Lymphoma.

Introduction: Annual surveillance CT scans and bone marrow (BM) biopsies are frequently performed in follow-up of patients (pts.) receiving high-dose therapy (HDT) with autologous stem cell rescue for recurrent follicular lymphoma (FL). The impact of this active surveillance strategy on the outcome of a homogeneously treated population was analysed.

Methods: Ninety-nine consecutive pts. who received HDT (cyclophosphamide and total body irradiation) with autologous stem cell rescue for recurrent FL at St Bartholomew’s Hospital between February 1986 and October 1991 were included. The surveillance policy at that time included an annual CT scan and BM biopsy. Time to relapse, time to next treatment and overall survival (OS) from the time of HDT were calculated and compared, according to whether disease progression had been diagnosed on the basis of surveillance investigations or on clinical grounds. Seventy out of 99 pts. are evaluable, the remainder are not, due to: relapse or death within one year of HDT (20 pts.), follow-up elsewhere (8 pts.) or patient’s wishes (1 pt.). 86% of pts. who commenced surveillance actually had annual CT scans and BM biopsies until disease progression or death.

Results: After a median follow-up of 16.7 years (y), progression was documented in 35/70 pts. (50%). It was detected by surveillance alone in 14 pts. and clinically, in 21 pts. (40% and 60% of all recurrences, respectively). The commonest presentations of clinical relapse were peripheral lymphadenopathy (62%), pain (19%) and B-symptoms (10%). Surveillance relapses were diagnosed on the basis of a CT scan in 43% of pts., a BM biopsy in 36% and both in 14%. The median time from HDT to relapse was 2y for pts. with a clinically detected relapse and 2.8y in those with a surveillance relapse but the difference was not statistically significant (p=0.2). Treatment was started immediately in 13 of 21 pts. (62%) with a clinical relapse, contrasting with only one of 14 pts. (7%) diagnosed on surveillance. The main reasons for starting treatment were biopsy-proven transformation to diffuse large B-cell pathology (8 of the pts. with a clinical relapse and one with a surveillance relapse) and rapidly progressing lymphadenopathy. All other pts. were managed expectantly (observation). The median time from HDT until clinical progression was 8.2y in the 14 pts. whose relapse was detected by surveillance; 4 of the 14 have not yet developed signs or symptoms of recurrence. Thus, the clinical relapse rate in the cohort under surveillance is 44% (31/70). Six of the pts. with a surveillance relapse have still not started treatment (median follow-up: 13.8y; range: 1.9y–22.1y). Only 2 of the 21 pts. with a clinically detected relapse have not started treatment after 5.6y and 14.3y of follow-up. Thus, the median time from HDT to next treatment was significantly shorter for pts. with a clinical relapse (3.6y) in comparison with those in whom the recurrence was diagnosed by surveillance investigations (11.3y; p=0.0004). Median OS was 5.6y and 13.4y for pts. with a clinical and a surveillance relapse, respectively (p=0.03).

Conclusions: In pts. with FL, relapses diagnosed on the basis of surveillance investigations usually have an indolent course and frequently do not require treatment. In contrast, clinically detected relapses have a more aggressive clinical course and treatment is initiated immediately in the majority of cases. Thus, annual surveillance investigations do not help to identify pts. that require treatment and do not improve the outcome of this population. They should therefore be abandoned. Time to relapse based on data from pts. on annual surveillance should be interpreted with caution because of the poor correlation with time to next treatment and OS.