Page 166 - 2020_07-Haematologica-web

P. Strati et al.
retrospective analysis of 346 patients with advanced stage FL, in whom a PET-CT scan was performed prior to initi- ation of therapy, and analyze the impact of SUVmax on the quality of response and outcomes after frontline therapy.
Methods
Patient selection
This is a retrospective analysis of patients with stage III or IV FL, grades I, II, or IIIA, receiving frontline treatment at the MD Anderson Cancer Center (MDACC) between August, 2001 and April, 2014, and who had a pre-treatment PET-CT scan performed. Patients with histological diagnosis of FL grade IIIB or concurrent diffuse large B-cell lymphoma (DLBCL) were excluded.
The clinical and laboratory features were confirmed by review of the medical records. Frontline therapy was adminis- tered according to the previously described schedule.23-27 The FLIPI and FLIPI-2 scores were calculated as previously described.28,29 Lugano classification was used to define complete response.30,31 The study was approved by the Institutional Review Board of the MDACC and conducted in accordance with our institutional guidelines and the principles of the Declaration of Helsinki.
PET scan and SUVmax threshold selection
Baseline PET-CT scans were obtained at MDACC before ini- tiation of frontline therapy. After patients had fasted for at least 4-6 hours, blood glucose was measured and confirmed to be <140 mg/dL (<200 mg/dL for patients with diabetes) before injection of 333-407 MBq (9-11 mCi) of [18F]FDG. Emission scans were acquired at 2-3 minutes per field of view in the three- dimensional mode after a 60-minute uptake time (±10 minutes). CT non-contrast images were acquired in helical mode with 3.75-mm slices from the skull base through the midthigh. Commercially available iterative algorithms were used for image reconstruction. PET images were collected and transferred to commercially available software (MIMVista version 6.4.9; MIMVista Corporation, Cleveland, OH). SUVmax was calculated as previously described.32 All reports of pre-treatment scans and of scans performed to assess response to frontline therapy were reviewed by an oncologist with expertise in lymphoma.
Analyzing multiple single unit increments of SUVmax among all 346 patients included in the study, 18 showed the strongest association with progression-free survival (PFS) (hazard ratio [HR] 1.5, 95% confidence interval [CI]: 0.95-2.3, P=0.08), and was selected as cut-off for further analysis (Online Supplementary Table S1).
Statistical methods
Results
Patient baseline characteristics
Three-hundred and forty-six patients were included in the study, median SUVmax was 11 (range: 1.5-42), and 52 (15%) patients had a SUVmax >18. All 52 patients with SUVmax >18 had a biopsy of the most FDG-avid lymph node, and no histological evidence of transformation was observed. Baseline characteristics are shown in Table 1. On univariate analysis, factors associated with SUVmax >18 were male sex (67% vs. 52%, P=0.05), elevated β2-microglobulin (65% vs. 47%, P=0.02), elevated lactate dehydrogenase (LDH)(37% vs. 13%, P<0.001), presence of B symptoms (35% vs. 14%, P=0.01), and a large lymph node >6 cm (64% vs. 30%, P<0.001)(Table 1). On multi- variate analysis, a large lymph node ≥6 cm was the only factor maintaining its association with SUVmax >18 (odds ratio [OR] 2.7, 95% CI: 1.3-5.3, P=0.006)(Table 2).
Response to frontline therapy: complete response (CR)
One-hundred and fifty-one (44%) patients were treated with frontline rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP), and 195 (56%) with other therapies, including rituximab and bendamustine (BR) in 55 (16%) patients, rituximab and lenalidomide (R2) in 63 (18%), rituximab, fludarabine, mitoxantrone, and dexamethasone (R-FND) in 24 (7%), and single agent rit- uximab in 53 (15%) patients. Two-hundred and thirty- two (65%) patients received maintenance rituximab and 114 (33%) were observed after completion of frontline therapy. While no differences in the use of maintenance therapy were observed between the two groups (75% vs. 67%, P=0.20), a significantly higher proportion of patients with baseline pre-treatment SUVmax >18 were treated with R-CHOP versus non-anthracycline-based regimens (75% vs. 38%, P<0.001), so subsequent results were stratified by treatment arm.
Among 342 patients evaluable for response, 305 (89%) achieved CR; the CR rate was 91% for patients with SUVmax <18 and 80% for patients with SUVmax >18 (P=0.03). No association between SUVmax and CR rate was observed among patients treated with R-CHOP (89% for SUVmax > 18 vs. 88% for SUVmax < 18, P=1). However, SUVmax >18 significantly associated with a lower CR rate among patients treated with other frontline regimens (45% for SUVmax >18 vs. 92% for SUVmax <18) (P<0.001) (Figure 1). After excluding patients treated with single agent rituximab a trend for a lower CR rate among patients treated with other frontline regimens and SUVmax >18 was observed (80% vs. 94%, P=0.17); of interest, in this group, only five patients with SUVmax >18 were evalu- able for response, likely limiting achievement of statistical significance.
Progression-free survival (PFS) and PFS24
After a median follow-up of 94 months (95% CI: 88-100 months), median PFS was not reached, and a trend for decreased PFS was observed in patients with baseline SUVmax >18 compared to patients with baseline <18 (114 months vs. not reached, P=0.08). Baseline ≥18 did not associate with shorter median PFS among patients treated with frontline R-CHOP (114 months vs. 144 months, P=0.73), but it did associate with significantly shorter PFS among patients treated with other frontline regimens (77 months vs. not reached, P=0.02)(Figure 2). After excluding
Association with categorical variables was evaluated using χ2 or Fisher exact tests, or the Mann-Whitney test, as appropriate, and logistic regression was used for multivariate analysis. Only factors significant (P-value <0.05) on univariate analysis were included in multivariate models. PFS was defined as the time from the start of therapy to progression of disease, death, or last follow-up (whichever occurred first). Overall survival (OS) was defined as the time from the start of therapy to death or last fol- low-up. PFS and OS were calculated for all patients in the study and for subgroups of patients using Kaplan-Meier estimates and were compared between subgroups using the log-rank test. Multivariable Cox regression analysis was performed to assess the associations between patient characteristics and PFS or OS. A P-value of <0.05 (two-tailed) was considered statistically sig- nificant. Statistical analyses were completed using SPSS 21.
1908
haematologica | 2020; 105(7)