Interim FDG‐PET does not predict outcome in advanced‐stage Hodgkin lymphoma patients treated with BEACOPP

Abstract

A recent study by Borchmann et al (2018), published in Lancet, included 1945 patients with advanced-stage Hodgkin lymphoma whose therapy was adapted on the basis of interim fludeoxyglucose positron emission tomography (FDG-PET) results after 2 cycles of bleomycin, etoposide, doxorubicin, cyclo phosphamide, vincristine, procarbazine and prednisone in escalated doses (eBEACOPP). FDG-PET scans after 2 cycles of eBEACOPP were considered positive when residual activity at PET was present with an intensity higher than the mediastinal blood pool (i.e. Deauville score ≥3). Patients with positive interim FDG-PET results after 2 cycles of eBEACOPP treated with standard therapy (69 or 89 eBEACOPP) had a 5-year progression-free survival (PFS) of 89 7%, whereas patients with positive interim FDG-PET results treated with intensified therapy (89 eBEACOPP with additional rituximab) had a 5-year PFS of 88 1%. Patients with negative interim FDG-PET results after 2 cycles of eBEACOPP treated with standard therapy (69 or 89 eBEACOPP) had a 5-year PFS of 90 8%, whereas patients with negative interim FDG-PET results treated with de-escalated therapy (49 eBEACOPP) had a 5-year PFS of 92 2%. Borchmann et al (2018) recommended interim FDG-PET adapted therapy in order to reduce the number of therapy cycles and associated morbidity in patients with negative interim FDGPET results. However, this recommendation was not supported by the results of their study. Note that patients with positive and negative interim FDG-PET results had a similar outcome when treated with 69 or 89 BEACOPP (5-year PFS of 89 7% and 90 8%, respectively), and that there was no experimental arm in which treatment de-escalation was applied in patients with positive interim FDG-PET results. Furthermore, as noted by the authors themselves, so far there have been no other studies on treatment adaptation on the basis of interim FDG-PET in patients treated with BEACOPP (Borchmann et al, 2018). The lack of an experimental arm applying treatment de-escalation in patients with positive interim FDG-PET results in the study by Borchmann et al (2018) and the lack of a comparative historical cohort applying treatment de-escalation in patients with positive interim FDG-PET results underline that no conclusion, even not yet a presumption, can be made on whether treatment de-escalation is more effective in patients with negative versus those with positive interim FDG-PET results. The fact that patients with positive interim FDG-PET scans had a favourable prognosis after standard eBEACOPP therapy (5-year PFS 89 7%) even suggests that treatment de-escalation may also be feasible in this subgroup of patients. It is acknowledged that some retrospective studies (Gallamini et al, 2007, 2014; Agostinelli et al, 2016) have shown patients with advanced-stage Hodgkin lymphoma treated with doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) and positive interim FDG-PET results after 2 cycles of chemotherapy to have a dismal prognosis, making treatment de-escalation in FDGPET positive patients rather counter-intuitive. However, these studies (Gallamini et al, 2007, 2014; Agostinelli et al, 2016) suffered from a major methodological flaw, because in the vast majority of cases, treatment failure was determined by means of follow-up imaging, suggesting disease progression after completion of therapy, and histological confirmation of relapsed disease was scarcely performed. Considering the reported high false-positive rate of followup FDG-PET in Hodgkin lymphoma (El-Galaly et al, 2012), the positive predictive value of FDG-PET during ABVD treatment in predicting outcome in advanced-stage Hodgkin lymphoma may have been seriously overestimated by those studies (Gallamini et al, 2007, 2014; Agostinelli et al, 2016). Consequently, whether a positive FDG-PET result after 2 cycles of chemotherapy equals a dismal prognosis remains unproven. A second important issue to consider is that the histological substrate of a positive interim FDG-PET scan is actually unknown, because studies that performed histological verification of residual FDG-avid lesions in Hodgkin lymphoma are lacking. On the other hand, multiple studies in nonHodgkin lymphoma that performed tissue sampling of FDGavid lesions at PET scans performed during treatment showed very high false positive rates (Table I), the vast majority of which were due to therapy-induced inflammatory changes. These results can very likely be extrapolated to Hodgkin lymphoma. Moreover, only 0 1–1 0% of the pathological tissue in Hodgkin lymphoma consists of malignant Reed-Sternberg cells (Pileri et al, 2002) and virtually all FDG-avidity is caused by surrounding inflammatory cells. These theoretical underpinnings underline that relevant numbers of false-positive interim FDG-PET results in Hodgkin lymphoma are not unlikely, and that treatment de-escalation in Hodgkin lymphoma patients with positive interim FDGPET scans may be feasible as well. Correspondence