Search

Supplementary Methods, Tables, and Figures

Purpose:Tafasitamab, an Fc-modified, humanized, anti-CD19 monoclonal antibody, in combination with lenalidomide, demonstrated efficacy in transplant-ineligible patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL), in the single-arm, phase II L-MIND study (NCT02399085). RE-MIND, a retrospective observational study, generated a historic control for L-MIND to delineate the contribution of tafasitamab to the efficacy of the combination.Patients and Methods:Data were retrospectively collected from patients with R/R DLBCL treated with lenalidomide monotherapy for comparison with tafasitamab + lenalidomide–treated patients (L-MIND). Key eligibility criteria were aligned with L-MIND. Estimated propensity score–based Nearest Neighbor 1:1 Matching methodology balanced the cohorts for nine prespecified prognostic baseline covariates. The primary endpoint was investigator-assessed best overall response rate (ORR). Secondary endpoints included complete response (CR) rate, progression-free survival (PFS), and overall survival (OS).Results:Data from 490 patients going through lenalidomide monotherapy were collected; 140 qualified for matching with the L-MIND cohort. The primary analysis included 76 patients from each cohort who received a lenalidomide starting dose of 25 mg/day. Cohort baseline covariates were comparable. A significantly better ORR of 67.1% (95% confidence interval, 55.4–77.5) was observed for the combination therapy versus 34.2% (23.7–46.0) for lenalidomide monotherapy [odds ratio, 3.89 (1.90–8.14); P < 0.0001]. Higher CR rates were achieved with combination therapy compared with lenalidomide monotherapy [39.5% (28.4–51.4) vs. 13.2% (6.5–22.9)]. Survival endpoints favored combination therapy. Lenalidomide monotherapy outcomes were similar to previously published data.Conclusions:RE-MIND enabled the estimation of the additional treatment effect achieved by combining tafasitamab with lenalidomide in patients with R/R DLBCL.

Purpose: Tafasitamab, an Fc-modified, humanized, anti-CD19 monoclonal antibody, in combination with lenalidomide, demonstrated efficacy in transplant-ineligible patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL), in the single-arm, phase II L-MIND study (NCT02399085). RE-MIND, a retrospective observational study, generated a historic control for L-MIND to delineate the contribution of tafasitamab to the efficacy of the combination. Patients and Methods: Data were retrospectively collected from patients with R/R DLBCL treated with lenalidomide monotherapy for comparison with tafasitamab + lenalidomide–treated patients (L-MIND). Key eligibility criteria were aligned with L-MIND. Estimated propensity score–based Nearest Neighbor 1:1 Matching methodology balanced the cohorts for nine prespecified prognostic baseline covariates. The primary endpoint was investigator-assessed best overall response rate (ORR). Secondary endpoints included complete response (CR) rate, progression-free survival (PFS), and overall survival (OS). Results: Data from 490 patients going through lenalidomide monotherapy were collected; 140 qualified for matching with the L-MIND cohort. The primary analysis included 76 patients from each cohort who received a lenalidomide starting dose of 25 mg/day. Cohort baseline covariates were comparable. A significantly better ORR of 67.1% (95% confidence interval, 55.4–77.5) was observed for the combination therapy versus 34.2% (23.7–46.0) for lenalidomide monotherapy [odds ratio, 3.89 (1.90–8.14); P < 0.0001]. Higher CR rates were achieved with combination therapy compared with lenalidomide monotherapy [39.5% (28.4–51.4) vs. 13.2% (6.5–22.9)]. Survival endpoints favored combination therapy. Lenalidomide monotherapy outcomes were similar to previously published data. Conclusions: RE-MIND enabled the estimation of the additional treatment effect achieved by combining tafasitamab with lenalidomide in patients with R/R DLBCL.

We analyzed safety of NEPA (netupitant/palonosetron) and dexamethasone (NEPA+DEX) for the management of chemotherapy-induced nausea and vomiting (CINV) in classical Hodgkin’s lymphoma patients that experienced CINV with a prophylaxis with palonosetron (PALO + DEX). In a retrospective, monocentric, noncomparative study, we analyzed adverse events and CINV grading in patients who switched from PALO + DEX to NEPA + DEX. Among 32 patients treated with ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) during the study period, 47% did not properly control CINV with PALO + DEX and were shifted to NEPA + DEX. Among these, 53.3% properly controlled CINV is for all the remaining chemotherapy cycles. We did not observe an increase of adverse events after switching to NEPA. In our study, NEPA did not show drug–drug interaction with ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) chemotherapy agents and NEPA administration was well tolerated with mild and transient adverse events.

The International Prognostic Score (IPS) is the most commonly used risk stratification tool for patients with advanced Hodgkin lymphoma (HL). It incorporates seven clinical parameters independently associated with a poorer outcome: male sex, age, stage IV, hemoglobin level, white blood cell and lymphocyte counts, and albumin level. Since the development of the IPS, there have been significant advances in therapy and supportive care. Recent studies suggest that the IPS is less discriminating due to improved outcomes with ABVD therapy. The aim of the present study was to asses if classic prognostic factors maintain their prognostic meaning at the time of response-adapted treatment based on interim PET scans. We evaluated the prognostic significance of IPS in the 520 advanced stage HL patients enrolled in the PET-guided, HD0801 trial in which PET2-positive patients underwent a more intense treatment with an early stem-cell transplantation after 2 cycles of ABVD. We observed that in these patients, the IPS completely loses its prognostic value together with all the single parameters that contribute to the IPS. Furthermore, neutrophils, monocytes, lymphocytes, and the ratio among them also no longer had any predictive value. We believe that the substantial improvement in survival outcomes in PET2-positive patients treated with early autologous transplantation could explain the complete disappearance of the residual prognostic significance of the IPS.

Context: Tafasitamab (MOR208), a humanized anti-CD19 antibody, combined with lenalidomide (LEN) showed encouraging results in the Phase II, single-arm L-MIND study ( NCT02399085 ) of patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL) ineligible for autologous stem cell transplantation (ASCT). Objective: To generate a LEN monotherapy matched comparator cohort for L-MIND using patient-level data to isolate the tafasitamab contribution to the efficacy of the tafasitamab + LEN combination. Design: RE-MIND ( NCT04150328 ), a retrospective observational study, used a Nearest Neighbor ePS-based 1:1 Matching methodology to balance the LEN monotherapy cohort (observational data) and combination cohort (derived from L-MIND) for nine prespecified baseline covariates: age, Ann Arbor stage, last therapy refractoriness, number of prior therapy lines, primary refractoriness, prior ASCT, LDH, neutropenia, and anemia. The primary analysis set included matched patients who received a LEN starting dose of 25 mg/day (as in L-MIND) with sufficient follow-up. Patients: LEN monotherapy-treated patients from the US and EU who met the L-MIND-aligned eligibility criteria: adult patients with R/R DLBCL; 1–3 prior systemic therapies, including 1 CD20-targeting regimen; ASCT-ineligible. Baseline characteristics, LEN dosing, and patient outcomes were collected retrospectively from health records. Main outcome measures: The primary endpoint was investigator-assessed best objective response rate (ORR). Secondary endpoints included overall survival (OS) and complete response (CR) rates. Results: 490 patients were enrolled, of which 140 fulfilled the ePS matching criteria. Following matching, the primary analysis set included 76 patients from each cohort. Baseline characteristics were comparable. The primary endpoint was met. Best ORR was significantly higher in the combination cohort (67.1%) versus the monotherapy cohort (34.2%) (odds ratio 3.89; 95% confidence interval [CI]: 1.90–8.14; p Conclusions: Significantly better ORR, CR, and OS indicate potential synergistic effects of the tafasitamab + LEN combination in ASCT-ineligible R/R DLBCL. RE-MIND demonstrates the utility of real-world data in interpreting non-randomized trials.

8020 Background: Patients with R/R DLBCL ineligible for autologous stem cell transplant (ASCT) have a poor prognosis. In these patients, tafasitamab (anti-CD19 antibody) plus lenalidomide (LEN) has shown encouraging results in the open-label, single-arm, phase II L-MIND study (n = 81; NCT02399085). To evaluate the contribution of tafasitamab to the activity of this doublet, we conducted a global, real-world study of patients treated with LEN monotherapy (RE-MIND; NCT04150328). Here we present the primary analysis of a 1:1 patient-level matched comparison between the L-MIND and RE-MIND cohorts. Methods: Patients treated with LEN monotherapy for R/R DLBCL were enrolled in the observational, retrospective RE-MIND cohort. As in L-MIND, patients had 1–3 prior systemic therapies, including ≥1 CD20-targeting regimen; were aged ≥18 years; and were not eligible for ASCT. A 1:1 estimated propensity score (ePS) matching methodology ensured balancing of nine pre-specified baseline covariates. The primary analysis set, Matched Analysis Set 25 (MAS25), included patients who received a LEN starting dose of 25 mg/day. The primary endpoint was investigator-assessed best objective response rate (ORR). Key secondary endpoints included overall survival (OS) and complete response (CR) rate. Results: 490 patients were enrolled in RE-MIND across 58 centers in the US and Europe, of which 140 fulfilled the ePS matching criteria. The MAS25 included 76 patients each from the two cohorts. Baseline characteristics between cohorts were comparable. The primary endpoint was met with a significantly better ORR of 67.1% (95% CI: 55.4–77.5) for the L-MIND cohort versus 34.2% (95% CI: 23.7–46.0) for the RE-MIND cohort (odds ratio 3.89; 95% CI: 1.90–8.14; p < 0.0001). The CR rate was 39.5% (95% CI: 28.4–51.4) in the L-MIND cohort and 13.2% (95% CI: 6.5–22.9) in the RE-MIND cohort. A significant difference in OS favored the L-MIND cohort (HR = 0.499; 95% CI: 0.317–0.785). ORR and CR outcomes in the RE-MIND cohort were similar to the published literature for LEN monotherapy in R/R DLBCL. Conclusions: Significantly better ORR, CR and OS indicate potential synergistic effects of the tafasitamab + LEN combination in ASCT-ineligible R/R DLBCL. ePS-based 1:1 matching allows robust estimation of the treatment effect of tafasitamab when added to LEN. RE-MIND demonstrates the utility of real-world data in interpreting non-randomized trials. Clinical trial information: NCT04150328 .

Optimal treatment for transformed follicular lymphoma (tFL) is not fully defined. Clinical characteristics and treatments that impact on post-transformation outcome of 176 biopsy-proven tFL were analysed. Transformation occurred at initial diagnosis in 52% (Group 1) and after a FL diagnosis in 48% (Group 2). Five-year overall survival was 84% for Group 1 and 51% for Group 2 (P < 0·001). In Group 1, 5-year progression-free survival was superior after rituximab maintenance compared to observation only (94% vs. 53%, P = 0·024). In Group 2, an inverse trend was found between survival and both a higher number of pre-transformation treatment lines and a short time-to-transformation.

BACKGROUND In the setting of cHL, ABVD (adriamycin, bleomycin, vinblastine, dacarbazine) is the most widely used first line chemotherapeutic treatment and it is well known that this regimen is associated with a high emetic risk (HEC). Palonosetron (PALO) currently represents one of the most effective and implemented drug for CINV prevention, but after many ABVD cycles patients (pts) frequently need to add other antiemetic drugs to obtain a good control of their symptoms. Netupitant (NETU) is the NK1-RA (neurokinin receptor antagonist) component of NEPA, the first antiemetic drug available as oral fixed combination: NETU (300mg) + PALO (0.5mg). Both ABVD drugs and NETU are metabolized by cytochrome P-450 isoform 3A4 (CYP3A4), but respect to other NK1-receptor antagonist available, NETU has demonstrated to have no clinical relevant interaction with chemotherapy drugs like etoposide, cyclophosphamide and docetaxel. However, no data are currently available about the safety profile of NETU in the setting of ABVD treatment; for that reason we started the use of this drug as salvage therapy after PALO failure. METHODS We retrospectively analyzed the cHL pts treated with ABVD at our Center from September 2016 to January 2018. We used PALO + dexamethasone as first-line anti-CINV prophylaxis, while NEPA was introduced as salvage drug for those pts with inadequately controlled CINV. We collected data regarding demographics; diagnosis; planned chemotherapeutic treatment; performed chemotherapeutic treatment; acute, delayed and anticipatory CINV (before and after NEPA); laboratory findings including transaminases, creatinine and electrolytes (before and after NEPA); adverse reactions (before and after NEPA). The primary endpoint of the study was safety of NEPA in ABVD treated pts, while CINV control (no nausea or vomiting) was the secondary endpoint. NEPA-related safety data have been compared to the same data collected at the moment of the last previous PALO-containing regimen. RESULTS Among the 32 pts treated with ABVD during the study period, 13 (41%) received NEPA. Three pts were males and 10 females, and median age was 33 years (range 18-61). According to disease characteristics at diagnosis the planned ABVD administrations were 12 (6 cycles) in 9 pts, 8 (4 cycles) in 3 pts and 4 (2 cycles) in 1 pt. Nine pts completed the planned chemotherapy, 1 pt skipped the last cycle for personal decision, 3 pts are ongoing treatment at the time of analysis. Reasons for shift to NEPA are as follows: acute (grade 2) CINV alone in 3/13 pts; late (grade 2) CINV alone in 3/13 pts; combined acute (4 grade 2, 1 grade 1) and late (4 grade 2, 1 grade 1) CINV in 5/13 pts; combined anticipatory (grade 1), acute (grade 1) and late (grade 2) CINV in 2/13 pts. NEPA was started after a median of 4 ABVD administrations (range 1-10). Globally 53 NEPA administrations were delivered during subsequent cycles (median number of 3 NEPA administrations for each pt, range 1-11). With regard to the primary endpoint, the observed adverse events are listed in Table 1. With regard to the secondary endpoint, anticipatory, acute and delayed CINV were detected in 15%, 77%, 77%, of PALO pts and 15%, 46% and 15% of NEPA pts, respectively (see Table 2) CONCLUSION In our experience NETU did not show drug-drug interaction with ABVD chemotherapy agents, and NEPA administration was globally well tolerated with mild and transient adverse events. Furthermore, in cHL ABVD treated pts who experienced nausea and/or vomiting after failure of PALO + dexamethasone antiemetic prophylaxis, NEPA has demonstrated to be effective in CINV control. In 4 out of 13 cases, after an initial improvement in CINV control, pts subsequently required to shift to anti-CINV third line treatments. On the other hand, the 9 pts who continued on NEPA administration could experience an optimal CINV control immediately after the first administration. At our knowledge no data have been published regarding NEPA toxicities in the ABVD setting. Our safety and efficacy data come from a real life experience of consecutive pts treated homogeneously at a single center and would suggest the use of NEPA as primary anti-CINV prophylaxis in previously untreated cHL pts. Disclosures Rusconi: Celgene: Research Funding.

Introduction: Transformation of Marginal Zone Lymphoma (MZL) into an aggressive histology is uncommon phenomenon that can occur at any time after diagnosis and is expected to have a detrimental impact on prognosis. Biological and clinical knowledge on transformed Marginal Zone Lymphoma (tMZL) is poor and no standard treatment is established in the Rituximab era for these patients (pts). We retrospectively analyzed all consecutive biopsy proven tMZL in two Italian Hematological Divisions from 2002 to 2014 and we focused on post-transformation treatment and outcome. Methods: The dataset included 378 MZL pts diagnosed between 2002 and 2014 at Division of Hematology in Pavia and in Milan (Niguarda Hospital): 204 pts (53.9%) by extranodal MALT lymphomas, 113 pts (29.8%) by splenic MZL, and 61 pts (16.3%) by nodal MZL. Histological transformation (HT) was defined as transformation into an aggressive lymphoma at any time from previous MZL diagnosis; only cases with biopsy confirmed HT were comprised in the present analysis, while cases with only clinical suspicion of transformation were not counted as tMZL. Results: HT was documented in 18 of the 378 pts (4.8%), 6.5% in nodal MZL, 7.0% in splenic MZL and 2.9% in MALT. Histology at transformation was Diffuse Large B-Cell lymphoma in all but one case; the remaining pt was diagnosed as high-grade B-cell lymphoma, unclassifiable. CD20 was negative only in one Rituximab-naïve pt. Median time from first diagnosis to HT was 31 months (range: 10-124) and median number of previous therapies was 1 (range 0-1); pts received first line therapy listed in table 1. Median age at transformation was 68 years (range: 46-85), M/F ratio was 0.8. In the tMZL population, first diagnosis was nodal MZL in 4 pts (22%), splenic MZL in 8 pts (45%) and MALT in 6 pts (33%). At first diagnosis of MZL, 72% of t-MZL pts had stage IV disease, 17% had B symptoms, 11% had elevated LDH and ECOG performance status was lower than 2 in all the cases. HCV serology was positive in 5/17 cases; HCV status was not available for one pt. At HT disease stage was III or IV in 14 pts (78%), B symptoms were present in 7 pts (39%), LDH and beta2microglobulin were both elevated in 7 pts (39%) and ECOG performance status was lower than 2 in all the cases. Pts received post-HT treatment listed in table 2. At time of analysis 6 pts died (33%), and the main cause of death was progressive disease. With a median post-transformation follow-up of 16.6 months (range: 2-98), the 2-years Progression-Free Survival (PFS) was 45,4 % and the 2-years Overall Survival (OS) was 56.75%. No correlation was found between the following characteristics and survival: MZL type at first diagnosis, stage, symptoms, LDH and ECOG at HT, number and types of pre-HT therapies. Conclusions: This large cohort confirms that HT is a relatively rare and early event in MZL. At present time, we did not identify any feature predictive of outcome for transformed MZL. Chemotherapy in combination with Rituximab showed to be an effective treatment for tMZL. Table 1 First line treatment N. of patients % Therapy strategy CVP 7 39 Anthracycline-containing regimen 2 11 Chlorambucil monotherapy 5 28 Splenectomy 1 5.5 H.pylori eradication 1 5.5 Watch and wait strategy 2 11 Rituximab incorporation Included in first line therapy 6 33 Maintenance 0 0 Response ORR (%) 67 CRR (%) 33 Table 2. Post-HT treatment N. of patients % Therapy strategy CHOP regimen 16 89 Platinum-containing regimen 1 5.5 Missing 1 5.5 Rituximab incorporation Included in first line therapy 17 94 Maintenance 0 0 ASCT consolidation 3 17 Response ORR (%) 61 CRR (%) 59 Disclosures Rusconi: Roche: Honoraria.

Abstract 4851 Background: Primary Mediastinal B-Cell Lymphoma (PMBL) is an uncommon disease, characterized by aggressive and invasive course but with a good prognosis after anthracycline-based chemotherapy. In the PET era the role of consolidation radiotherapy is under debate and, despite CD20 expression, the efficacy of Rituximab is still unclear. We retrospectively analyzed the outcome of 36 consecutive patients (pts) affected by PMBL treated in the last 10 years at our institution. We focused on anti-CD20 antibody efficacy when added to chemotherapy and on the role of autologous stem cell transplantation (ASCT) in PET positive pts after first-line treatment. Patients and methods: From June 2000 to March 2011 36 pts with biopsy proven PMBL referred to our institution. Median age was 35 years (range: 18–68); 21 pts (58%) were female and a mediastinal bulk at diagnosis was documented in 33 pts (92%). B-symptoms were reported in 16 cases (44%) and an extra-nodal involvement in 19 cases (53%). Age-adjusted IPI score was ≥ 2 in 12 pts (33%). For all patients first line treatment consisted in a third generation anthracycline-based chemotherapy (VACOP-B), with the addiction of 6 Rituximab doses in 15 pts (42%). Pts obtaining complete remission (CR) with negative PET after (R)-VACOP-B were consolidated by radiotherapy (RT), while pts in partial remission (PR) with residual FDG uptake underwent a second-line chemotherapy with 3 DHAP cycles followed by autologous stem cell transplantation (ASCT). Results: In the whole cohort, after first-line therapy overall response rate (ORR) was 97%, with a CR rate of 39%. RT was therefore performed in 14 PET-negative pts. 2/14 pts experienced early relapse and only one of them obtained a second CR after salvage therapy, while the non-responding patient died because of progressive disease. Twenty-one pts (58%) showed a residual FDG uptake after (R)-VACOP-B and underwent second-line therapy. Nineteen pts responding to second-line therapy achieved ASCT, while 2 pts progressed and died after salvage therapy. ORR after ASCT was 86% with a CR rate of 71%. Post-ASCT RT was performed in 10 pts, 7 CR and 3 PR; two PR pts converted to CR after RT. With a median follow-up of 66 months (range: 13–142) 2-year overall survival (OS) and progression free survival (PFS) were respectively 94% and 89%. Among the 15 pts receiving first-line chemotherapy containing Rituximab, ORR after R-VACOP-B was 93% with a CR rate of 40%. RT was therefore performed in 6 PET-negative pts. 1/6 pts experienced early relapse and died of progressive disease. One patient showed progressive disease after R-VACOP-B and underwent second-line therapy with ASCT, obtaining CR. Eight pts (53%) showed a residual FDG-uptake after R-VACOP-B and underwent second-line therapy and ASCT. ORR and CR rate after ASCT were 100% and 75% respectively. Two pts in PR after ASCT converted to CR after RT. Among the 21 pts receiving chemotherapy without Rituximab, ORR after VACOP-B was 100% and CR rate was 38%. RT was therefore performed in 8 PET-negative pts; one of them experienced early relapse and obtained a second CR after salvage therapy. Thirteen pts (62%) showed a residual FDG-uptake after VACOP-B and underwent second-line therapy. Eleven pts responding to second-line therapy achieved ASCT, while 2 pts progressed and died after salvage therapy. ORR and CR rate after ASCT were 77% and 69% respectively. No statistically significant difference in ORR, CR rate, OS and PFS (Figure 1) was found between pts treated with Rituximab plus chemotherapy and pts treated with chemotherapy alone. Conclusions: These data substantially confirm the satisfactory outcome of PMBL, with a 2-year OS and PFS of 94% and 89% for the entire cohort. We registered a residual FDG uptake after first line chemotherapy in a proportion of pts higher than expected (58%). This subgroup of pts clearly take advantage from second line chemotherapy followed by ASCT, obtaining a CR rate of 71%. The addiction of Rituximab to first line chemotherapy instead does not seem to improve PMBL pts outcome in this small and retrospectively analyzed population. The role of immunotherapy in this rare lymphoma subtype and the chance to safely avoid RT consolidation in PET negative pts need to be further investigated in wider prospective trial. Disclosures: No relevant conflicts of interest to declare.

Abstract 1614 Background: Few data are available on the role of fluoro-deoxy-glucose positron emission tomography (FDG-PET) in T-cell non Hodgkin Lymphoma (NHL). We compared standard contrast-enhanced computerized tomography (CECT) and FDG-PET for initial staging and restaging after treatment in T-cell NHL. Methods: We reviewed 58 FDG-PET scans performed in 29 patients (pts) affected by T-cell NHL and we focused on the 42 cases for which a simultaneous standard CECT was available for comparison. The specific T-cell histology was Peripheral T-Cell Lymphoma Not Otherwise Specified in 12 pts, Angioimmunoblastic T-Cell Lymphoma in 4 pts, Anaplastic Large Cell Lymphoma in 10 pts, other T-NHL hystologies in 3 pts. Twelve pts were males and median age at diagnosis was 57 years (range: 25–77). Baseline FDG-PET scans were 20, while FDG-PET performed for disease reassessment were 22. Results: FDG-PET performed at baseline was positive in all cases. In 6 cases (30%) FDG-PET identified fewer disease sites than CECT, while in 14 cases (70%) FDG-PET identified additional disease sites. New sites identified by FDG-PET were: other nodal sites (11 pts), spleen (2 pts), nasopharynx (1 pt) and testicles (1 pt). Different disease extent evaluation according to FDG-PET modified clinical stage in 10 pts (50%): one pt was downstaged and 9 pts were upstaged. FDG-PET-based stage was not considered for therapeutic decision. FDG-PET scans performed at restaging were negative in 16 cases (72%) and positive in the remaining 6 cases (28%). In 13/16 (81%) negative cases CECT was concordant and showed a complete remission, while in the remaining 3 cases CECT showed a residual nodal disease. In the 6 FDG-PET positive cases, CECT was as follows: abnormal in the same site of FDG-PET in 1 case, abnormal in fewer sites than FDG-PET in 3 cases and abnormal in more sites than FDG-PET in 2 cases. With a median follow-up of 32 months, the median overall survival (OS) of the 6 positive FDG-PET cases at restaging was 26 months. With a median follow-up of 23 months, the median OS of the 16 negative FDG-PET cases at restaging was not reached. Conclusion: FDG-PET can be useful in the initial staging of T-cell NHL, since it can allow a more accurate disease extension evaluation: in our experience it could identify additional sites, both nodal and extra-nodal, in 70% of pts. A change in clinical stage according to FDG-PET was observed in 50% of cases, with an upstaging in the majority of pts. We could confirm the high sensitivity of baseline FDG-PET in T-cell NHL. A high level of concordance was observed between PET and CECT at restaging. However, independently on CECT result, the worse OS of FDG-PET positive cases confirms the role of FDG-PET as prognostic factor for survival. Larger prospective trials are needed to confirm the utility of FDG-PET in baseline staging, reassessment after chemotherapy and treatment planning. Disclosures: No relevant conflicts of interest to declare.

Abstract 3120 Introduction Fluorodeoxyglucose positron emission tomography (FDG-PET) is a functional imaging tool routinely used for staging and response assessment in aggressive lymphomas. Even if mantle cell lymphoma (MCL) is considered FDG avid, current data on the use of PET in this histotype are scant. We retrospectively analysed 43 PET scans in 22 MCL patients (pts) and compared the results to standard contrast-enhanced computerized tomography (CT) in order to define PET sensitivity and specificity in this rare lymphoma. Patients and methods Twenty-two pts with biopsy proven diagnosis of MCL underwent PET in different phases of disease for a total of 43 scans. PET scans were co-registrated with a simultaneous CT and compared to standard contrast-enhanced CT performed within 4 weeks. Nineteen PET (44.2%) were performed for disease staging at diagnosis or relapse, while 24 (55.8%) for response assessment. Negative PET were included in the analysis for response assessment only if a previous positive scan was available. Results PET and standard CT results were concordant in 16/43 cases (37.2%). Discrepancies were found in 27/43 cases (62.8%) and were grouped in four categories: false negative PET (5 cases), PET positive with a disease extension minor or major than CT (7 and 9 cases, respectively) and PET positive with negative CT (6 cases). At staging PET was concordant with CT in 4/19 (21%) cases. In 5/19 cases (26.3%) PET was negative in all disease sites documented by standard CT. In the remaining 10 cases (52.7%) PET was positive but discordant with disease extension evaluation by CT, determining upstaging and downstaging in 5 scans (26.3%) each. Involvement of gastro-intestinal tract was detected only in 4/19 cases (21%). PET performed for response assessment was concordant with CT in 12/24 cases (50%). In 6/12 concordant cases (25%) both CT and PET were negative; in the remaining 6/12 cases (25%) PET was positive and disease extension evaluation was concordant with standard CT. Six out of 12 discordant cases (25%) presented different disease extension evaluation; in the other six cases standard CT was negative while PET documented minimal residual disease (MRD). PET sensitivity and specificity in the entire cohort were 86.5% and 100%, respectively; positive predictive value was 100%, while negative predictive value was 54.5%. No difference in PET sensitivity was found in nodal and extra-nodal sites. No statistical significant correlation was found between PET sensitivity and proliferation index; nevertheless, no pts with Mib-1>50% presented either false negative PET scan or disease extension underestimated by functional imaging. Conclusion This analysis shows a PET sensitivity (86.5%) lower than expected if compared with other aggressive lymphomas. High specificity registered in this cohort is likely to be ascribed to simultaneous CT co-registration, that allows identification of false positive cases. A false negative PET scan at initial staging was registered in 21% of cases. A gastro-intestinal tract involvement, expected to be near universal in advanced stage MCL, was detected by functional imaging only in 21% of pts at disease onset or relapse. Either downstaging or upstaging of disease burden seldom result in different therapeutic approach since localized MCL is a rare entity. Minimal residual disease, detected by functional imaging in half of standard CT negative pts at response assessment, has uncertain role in MCL, since the disease remains incurable and there is no clear evidence that MRD positivity requires treatment. All these findings advise against utilization of functional imaging for routine management of MCL. Considering heterogeneity of FDG-uptake in MCL, PET is likely to be more reliable in pts presenting high proliferation index. Moreover, current data suggest a prognostic value of baseline FDG-avidity. Further investigations are needed to confirm these preliminary data in larger prospective trial, separating MCL from other aggressive histotypes. Disclosures: No relevant conflicts of interest to declare.