Multicenter studies are needed to demonstrate the clinical potential value of radiomics as a prognostic tool. However, variability in scanner models, acquisition protocols and reconstruction settings are unavoidable and radiomic features are notoriously sensitive to these factors, which hinders pooling them in a statistical analysis. A statistical harmonization method called ComBat was developed to deal with the "batch effect" in gene expression microarray data and was used in radiomics studies to deal with the "center-effect". Our goal was to evaluate modifications in ComBat allowing for more flexibility in choosing a reference and improving robustness of the estimation. Two modified ComBat versions were evaluated: M-ComBat allows to transform all features distributions to a chosen reference, instead of the overall mean, providing more flexibility. B-ComBat adds bootstrap and Monte Carlo for improved robustness in the estimation. BM-ComBat combines both modifications. The four versions were compared regarding their ability to harmonize features in a multicenter context in two different clinical datasets. The first contains 119 locally advanced cervical cancer patients from 3 centers, with magnetic resonance imaging and positron emission tomography imaging. In that case ComBat was applied with 3 labels corresponding to each center. The second one contains 98 locally advanced laryngeal cancer patients from 5 centers with contrast-enhanced computed tomography. In that specific case, because imaging settings were highly heterogeneous even within each of the five centers, unsupervised clustering was used to determine two labels for applying ComBat. The impact of each harmonization was evaluated through three different machine learning pipelines for the modelling step in predicting the clinical outcomes, across two performance metrics (balanced accuracy and Matthews correlation coefficient). Before harmonization, almost all radiomic features had significantly different distributions between labels. These differences were successfully removed with all ComBat versions. The predictive ability of the radiomic models was always improved with harmonization and the improved ComBat provided the best results. This was observed consistently in both datasets, through all machine learning pipelines and performance metrics. The proposed modifications allow for more flexibility and robustness in the estimation. They also slightly but consistently improve the predictive power of resulting radiomic models.

Autologous stem cell transplantation (ASCT) as first-line therapy for follicular lymphoma (FL) remains controversial. The multicenter study randomized 172 patients with untreated FL for either immunochemotherapy or high-dose therapy (HDT) followed by purged ASCT. Conditioning was performed with total body irradiation (TBI) and cyclophosphamide. The 9-year overall survival (OS) was similar in the HDT and conventional chemotherapy groups (76% and 80%, respectively). The 9-year progression-free survival (PFS) was higher in the ASCT than the chemotherapy group (64% vs 39%; P ‫؍‬ .004). A PFS plateau was observed in the HDT group after 7 years. On multivariate analysis, OS and PFS were independently affected by the performance status score, the number of nodal areas involved, and the treatment group. Secondary malignancies were more frequent in the HDT than in the chemotherapy group (6 secondary myelodysplastic syndrome/acute myeloid leukemia and 6 second solid tumor cancers vs IntroductionFollicular lymphomas (FLs) are a subgroup of B-cell nonHodgkin lymphomas (NHLs) accounting for 15% to 30% of newly diagnosed lymphomas. [1][2][3] The natural history of this disease is characterized by a long survival that contrasts with systematic relapses. Many chemotherapy regimens have been used for treatment but have not improved long-term outcome, which depends on initial prognostic factors such as the Follicular Lymphoma International Prognostic Index (FLIPI), 4 response to first-line therapy, and minimal residual disease indicated by positivity of the bcl2 rearrangement on analysis of the polymerase chain reaction (PCR). Although new therapeutic approaches, including purine analogs and anti-CD20 monoclonal antibodies, have shown impressive response rates and prolonged progression-free survival (PFS), their effects on overall survival (OS) have yet to be confirmed. The major improvement in the management of FLs has been achieved by the use of the anti-CD20 monoclonal antibody rituximab, which improved OS in combination with chemotherapy in all 4 prospective, randomized studies published to date. [5][6][7][8][9] Before the era of monoclonal antibodies, autologous stem cell transplantation (ASCT) was evaluated as an alternative approach to standard chemotherapy. Numerous studies have shown encouraging results for patients with relapsed follicular NHL, [10][11][12][13] 10,15,16 The fourth study, from the Gruppo Italiano Trapianto di Midollo Osseo (GITMO), compared chemotherapy to ASCT with rituximab in both arms. 17 In 2005 we published the results of the Groupe Ouest-Est d'Etude des Leucémies et Autres Maladies du Sang (GOELAMS) 064 trial, showing a higher overall response rate and longer median event-free survival (EFS) in the high-dose therapy (HDT) group then the conventional chemotherapy group after a median follow-up of 5 years, with no difference in OS but a higher rate of secondary malignancies in the HDT group. 18 We report here the final results of this trial, after an extended follow-up period of 108 months (9 years...

Doxorubicin-based immunochemotherapy, with interferon, has been shown to improve survival in patients with advanced follicular lymphoma. High-dose chemotherapy with stem-cell support is effective in follicular lymphoma in relapse but remains controversial as a first-line therapy. In a randomized study using a purged autologous stem-cell support, we compared these 2 approaches in patients with advanced follicular lymphoma. Newly diagnosed advanced follicular lymphoma patients (172 patients) were randomly assigned either to an immunochemotherapy regimen (cyclophosphamide, doxorubicin, teniposide, prednisone, and interferon) or to a high-dose therapy followed by purged autologous stem-cell transplantation. Compared with the patients who received chemotherapy and interferon, patients treated with high-dose therapy had a higher response rate (69% vs 81%, P ‫؍‬ .045) and a longer median event-free survival (not reached vs 45 months). This did not translate into a better survival rate due to an excess of secondary malignancies after transplantation. The Follicular Lymphoma Prognostic Index identified a subgroup of patients with a significantly higher event-free survival rate after highdose therapy. Autologous stem-cell transplantation cannot be considered as the standard first-line treatment of follicular lymphoma for patients younger than 60 years old with a high tumor burden. (Blood. 2005;105:3817-3823)

No abstract