LAUSR

Dear Editor, With interest, we read the article by Moskowitz et al. [1] that was recently published in Blood. Their study included 65 patients with relapsed or refractory Hodgkin lymphoma who were transplant eligible and who were treated with brentuximab with or without augmented ifosfamide, carboplatin, and etoposide (ICE), followed by high-dose therapy and autologous stem cell transplantation (ASCT). The aim of this study was to assess the value of pretherapyestablished risk factors and several new risk classification biomarkers such as pretherapy cytokine levels, pretherapy FDGPET metrics, and pre-transplant FDG-PET status. In the multivariate model, primary refractory disease to frontline therapy (P = 0.003) and baseline total metabolic volume (bTMV; P < 0.001) were reported to be independently predictive of event-free survival (EFS). Patients with low MTV/negative pre-transplant PET (n = 41), low MTV/positive pretransplant PET (n = 7), high MTV/negative pre-transplant PET (n = 8), and high MTV/positive pre-transplant PET (n = 3) had 3-year EFS rates of 93, 86, 38, and 0%, respectively (Figure 4c in the article of Moskowitz et al. [1]). Moskowitz et al. [1] concluded bMTV and pre-transplant FDG-PET results to be independently predictive and claimed bMTV to improve the predictive power of pre-transplant FDG-PET. However, we strongly disagree with their conclusion, because this statement was simply not supported by their own data. Note that pre-transplant FDG-PET results were not even included in their multivariate analysis (Table 4 in the article of Moskowitz et al. [1]), and that in this analysis, only refractory disease to frontline therapy and bMTVwere reported to be independently predictive of EFS. Furthermore, we do not agree that bMTV improves the predictive value of pre-transplant FDGPET. Note that both patients with negative (3-year EFS 93%) as well as positive (3-year EFS 86%) pre-transplant FDG-PET had a very good prognosis as long as the bMTVwas low. On the other hand, both patients with negative (3-year EFS of 38%) as well as positive (3-year EFS of 0%) pre-transplant FDG-PET had a dismal prognosis, as long as the bMTVwas high. Finally, of the 12 relapses that occurred in their cohort, 8 (67%) had a negative pre-transplant FDG-PET. These findings seriously question the statement that pre-transplant FDG-PET status has clinically relevant additional value to the baseline risk stratification, which is unlikely to be statistically significant contrasting the claims of Moskowitz et al. [1]. The fact that patients’ prognosis depends on the baseline risk rather than the FDG-PET response status is well known from clinical practice. Patients with early-stage Hodgkin lymphoma treated in the frontline setting have a very good prognosis, even when they have a positive FDG-PET result during treatment [2, 3], similar to the results by Moskowitz et al. [1] during second-line therapy, in which those with low bMTV and positive pre-transplant FDG-PET scans had a very good prognosis (3-year EFS 86%). On the other hand, several studies have shown advanced-stage Hodgkin lymphoma patients with negative FDG-PET results during treatment to have a less favorable prognosis, despite the favorable response according to FDG-PET [4–7]. Particularly, the results presented in the study by Moskowitz et al. [1] are striking, because the group of patients with a high bMTV and a negative pre-transplant FDG-PET had a dismal 3year EFS of only 38%, in spite of the fact that FDG-PET suggested a very good response to therapy (from high * Hugo J.A. Adams [email protected]