REPLY: We thank Meignan et al. for their interest in our letter, in which we document the limitations and low necessity of interim 18F-FDG PET/CT imaging in lymphoma. Although Meignan et al. seem to agree with us that interim 18F-FDG PET/CT has low clinical value in non-Hodgkin lymphoma (1,2), they disagree on the value of interim 18F-FDG PET/CT in Hodgkin lymphoma, which we will therefore discuss in this reply.
Hodgkin lymphoma is usually divided into early- and advanced-stage disease, which are treated differently and have a different prognosis. Studies have shown that the value of interim 18F-FDG PET/CT for predicting outcome is not homogeneous in these different disease entities. In early-stage Hodgkin lymphoma, the value of interim 18F-FDG PET/CT can be considered low: patients with positive interim 18F-FDG PET/CT findings have been reported to have a generally good progression-free survival (range, 30%–100%) and an excellent overall survival (range, 85.2%–100%) after standard, nonintensified, therapies, with most studies estimating long-term progression-free survival of higher than 80% (3). Consequently, it has to be concluded that most patients with positive interim 18F-FDG PET/CT findings remain disease-free after finishing nonintensified treatment and that second- and third-line therapies can cure most patients in whom first-line therapy fails. This seriously questions whether early treatment intensification based on interim 18F-FDG PET/CT results is justified. Results from the randomized EORTC/LYSA/FIL H10 trial (4) showed that interim 18F-FDG PET/CT–positive patients treated with intensified regimens (2 cycles of ABVD [doxorubicin, bleomycin, vinblastine, dacarbazine] plus 2 cycles of BEACOPP-escalated therapy [bleomycin, etoposide, adriamycin, cyclophosphamide, vincristine, procarbazine, and prednisone] plus radiation therapy) had a better 5-y progression-free survival than those treated with standard therapy (3 cycles of ABVD plus radiation therapy) (91% vs. 77%), but overall survival did not significantly differ between these two groups (4), supporting our aforementioned statement. On the other hand, although the relapse rate of early-stage Hodgkin lymphoma patients with negative interim 18F-FDG PET/CT results treated with standard therapies is low from an absolute point of view (5–7), it is actually high considering the generally good prognosis of these patients (long-term progression-free survival, ∼93% (8)), which underlines that a negative interim 18F-FDG PET/CT result cannot reliably exclude residual disease (9). Although randomized studies applying interim 18F-FDG PET/CT–based treatment deescalation (5,6) have shown that interim 18F-FDG PET/CT–negative patients have a generally good outcome after being treated with less intensive therapies, this is more likely a reflection of the generally good prognosis of the disease rather than of the negative predictive value of interim 18F-FDG PET/CT (10). From a relative point of view, disease relapse occurs much more frequently in patients treated with deescalated therapy than in those who continue standard therapy despite negative interim 18F-FDG PET/CT results (hazard ratios of up to 9.36 have been reported (5,6,10)). Considering the low positive and negative predictive values, it remains questionable whether an interim 18F-FDG PET/CT–based therapeutic approach is justified in early-stage Hodgkin lymphoma. Not the least of the reasons for this question is the fact that in this disease several other biomarkers, inexpensive and easily available (e.g., the risk models of the European Organisation for Research and Treatment of Cancer, German Hodgkin Study Group, and National Comprehensive Cancer Network (11)), have shown prognostic value equaling that of interim 18F-FDG PET/CT and may be a better surrogate for risk-adapted trials. 18F-FDG PET/CT scans are expensive, expose patients to potentially harmful ionizing radiation, are uncomfortable for patients, and are not available in all institutions (particularly in non-Western countries). Therefore, it is not unlikely that interim 18F-FDG PET is useless in early-stage Hodgkin lymphoma.
In advanced-stage Hodgkin lymphoma, results on the predictive value of interim 18F-FDG PET/CT are less consistent. Two studies by Gallamini et al. (12,13) reported interim 18F-FDG PET/CT to have excellent positive and negative predictive values. Patients with positive interim 18F-FDG PET/CT results had a dismal progression-free survival of 12.8% in one study and 28% in the other, whereas patients with negative interim 18F-FDG PET/CT results had an excellent progression-free survival of 95% in both studies after finishing standard ABVD therapy (12,13). However, both studies had a major methodologic flaw: histologic confirmation was available for only a small minority of cases of relapse, with relapse being documented by follow-up imaging in most cases (12–15). Posttreatment and follow-up 18F-FDG PET/CT studies have a strikingly high number of false-positive results, as has been reported for several lymphoma subtypes (16–19), including Hodgkin lymphoma (20). Consequently, the studies by Gallamini et al. (12,13) are methodologically seriously biased. The predictive value of interim 18F-FDG PET/CT was generally lower in other comparable studies (21), and 2 recent studies (22,23) including advanced-stage lymphoma as part of their patient population showed interim 18F-FDG PET/CT to have minor or no value in predicting prognosis. Three recent studies (24–26) on treatment intensification in interim 18F-FDG PET/CT–positive patients were published, all lacking a randomized control arm with nonintensified treatments. Consequently, the true benefit of treatment intensification in these patients could not be assessed. In addition, comparisons with historical studies that suffered from inadequate methodology and heterogeneous results are futile (21). We individually criticized all 3 of these studies for these issues (27–29). On the other hand, multiple, recently published, large-scale studies (14,24–26) unambiguously showed that (in contrast to the studies by Gallamini et al. (12,13)) a high proportion of the large group of patients with negative interim 18F-FDG PET/CT results develops disease relapse during follow-up and that, therefore, a negative interim 18F-FDG PET/CT result cannot exclude residual disease. In other words, most relapses occur after a negative interim 18F-FDG PET/CT result (14,24–26). One should consider this concern when interpreting the interim results of the study by Casasnovas et al. (currently published only in abstract form (30)) on treatment deescalation in 18F-FDG PET/CT–negative patients, which is the only randomized study yet available claiming that 18F-FDG PET/CT–based treatment deescalation is feasible (excepting a study by Johnson et al. (24), who made a minor change in treatment—omitting bleomycin in interim 18F-FDG PET/CT–negative patients—without finding a significant increase in relapse rate).
In conclusion, interim 18F-FDG PET/CT is not justified in early-stage Hodgkin lymphoma. The value of positive interim 18F-FDG PET/CT results in advanced-stage Hodgkin lymphoma is not well established because of methodologic issues in historical studies and the lack of a control or randomization arm in recent 18F-FDG PET/CT–adapted trials. On the other hand, most disease relapses in advanced-stage Hodgkin lymphoma occur in the large group of interim 18F-FDG PET/CT–negative patients, underlining that residual disease cannot be excluded and that treatment deescalation in these patients is highly questionable. Except for the interim results of the AHL2011 LYSA trial by Casasnovas et al. (30), there are no data confirming that treatment deescalation in interim 18F-FDG PET/CT–negative advanced-stage Hodgkin lymphoma is feasible. Therefore, there is currently no convincing evidence to support use of interim 18F-FDG PET/CT, either for prognostication or for treatment adaptation, in the routine clinical care of patients with early- or advanced-stage Hodgkin lymphoma.
Footnotes
Published online Mar. 9, 2017.
- © 2017 by the Society of Nuclear Medicine and Molecular Imaging.