Portal de Periódicos da CAPES

A polymorphism in FASL is associated with rituximab response in follicular lymphoma patients

2016; Wiley; Volume: 91; Issue: 6 Linguagem: Inglês

10.1002/ajh.24354

1096-8652

R. Gutierrez-Civicos, Ana María Hurtado, Daniel Torres‐Moreno, José Javier Sánchez‐Blanco, Ignacio Español, Luciano Consuegra‐Sánchez, Elena Pérez‐Ceballos, María Dolores Gutiérrez‐Meca, Andrés Jerez, Pablo Conesa‐Zamora,

Immune Cell Function and Interaction

To the Editor: Follicular lymphoma (FL) is an incurable non-Hodgkin lymphoma (NHL), characterized by a typically indolent clinical course presenting multiple episodes of relapse 1. Despite the fact that rituximab in combination with chemotherapy has proven to increase FL progression free and overall survival, there is considerable inter-individual variability in the clinical response to this anti-CD20 monoclonal antibody (MAb) 2. Apoptosis induction seems to be a process involved in the mechanism of action of MAbs 2-4 and several studies have reported the association of polymorphisms in genes coding for apoptotic extrinsic pathway proteins, such as death receptors and their ligands, with susceptibility to lymphomas or immunotherapy response 5, 6. However, there are no studies analyzing the relationship between polymorphisms in genes coding for extrinsic apoptosis inducers and rituximab response. In this work, we aimed to study whether rs763110 (−844C > T) FASL polymorphism, located in a putative binding domain for a transcription factor 3, might influence the rituximab response in FL patients after first-line treatment. The cohort included 126 FL patients recruited between January 2002 and May 2014 from the Hematology outpatient clinics from three hospitals from Southeast of Spain. The study was approved by the respective hospital ethics committees. All patients were treated with rituximab alone or in combination with the following first line schemes: R-CHOP-type (rituximab, cyclophosphamide, with or without doxorubicin or mitoxantrone, vincristine, prednisone), R-CHOP + R-type (rituximab, cyclophosphamide, with or without doxorubicin, vincristine, prednisone and two additional infusions of rituximab), R-FMC (rituximab, fludarabine, mitoxantrone, cyclophosphamide), R-FC (rituximab, fludarabine, cyclophosphamide), R-FMD (rituximab, fludarabine, mitoxantrone, dexamethasone), rituximab-bendamustine, and rituximab-clorambucile. Clinical response was evaluated 1–2 months after the last treatment course by means of physical examination and computed tomography (CT) or positron emission tomography (PET)-CT scans. Responses were scored according to the International Working Group Consensus Revised as complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). With the purpose of the analyses, patients were grouped as "partial/non-responders" (including PR, SD, and PD) and "complete responders" (CR). Peripheral blood samples were collected and DNA was extracted using the QIAamp DNA Mini Kit and the QIAcube automatic nucleic acid extractor (Qiagen, Venlo, the Netherlands) according to the purveyor's instructions. FASL rs763110 genotypes were determined by fluorescence resonance energy transfer competitive allele-specific PCR using the Kbioscience chemistry kit (Hertfordshire, UK). Baseline characteristics of the study patients and their correlation with response are shown in Supporting Information 1. Median age at diagnosis was 56.1 ± 15.1 years and 56.3% were male patients. Lactate dehydrogenase, β2-microglobulin (β2M), and hemoglobin levels were normal in 31.7%, 18.3%, and 70.6%, respectively. Most of patients had stage III–IV (76.2%) and grade 1–2 (67.5%) at diagnosis. The average number of rituximab infusions administered was 6.4 ± 1.2. Clinical responses achieved were a CR in 91 patients (72.2%) PR in 30 (23.8%), SD in 4 (3.2%), and 1 in PD (0.8%). Subgrouped as described above, 91 patients were considered "complete responders" and 35 "partial/non-responders." No significant differences were observed for the baseline features between "complete responders" and "partial/non-responders," except for male gender which was associated with a worse rituximab response (P = 0.012). Genotype frequencies fulfil Hardy–Weinberg disequilibrium and were similar to those found in public databases and prior studies on Caucasian population (http://www.ncbi.nlm.nih.gov/projects/SNP) 3. No significant differences between the antineoplastic schemes applied and the treatment response or the genotype distributions were observed (Supporting Information 1, Table 1). Patients with complete response were more frequently FASL CC than TC or TT carriers (70.5%, 42.4%, 47.4%, respectively). TC/TT genotypes were more often found in "partial/non-responders" (85.7%) than in "complete responders" (65.9%) reaching statistical significance (P = 0.028) (Table 1). When only R-CHOP patients were considered (n = 95) similar results were still found (TC/TT: 92.3% vs. 63.8%, P = 0.004). Supporting Information 2 shows the result of a non-adjusted binary logistic regression analysis. Variables considered as covariates in the enter-method multivariable binary logistic regression model were those outside the range (odds ratio (OR9) 0.66–1.50), irrespective of the P-value. We further selected those more relevant candidates to construct a parsimonious model and avoid overfitting. The variables female gender, stage III–IV, elevated β2M, and FASL rs763110 T allele were the significant predictor variables chosen to carry out the multivariable binary logistic regression model associated with the condition of "complete responder." We found that female gender (OR = 0.202, CI 95% 0.066–0.623, P = 0.005) and T allele in rs763110 (OR = 5.321, IC 95% 1.357–20.860, P = 0.016) were independent predictors of response. Neither stage III–IV (OR = 2.205, CI 95% 0.600–8.107, P = 0.234) nor β2M elevated (OR = 2.990, CI 95% 0.729–12.269, P = 0.128) were predictors of response. This final model showed a good calibration with a Hosmer–Lemeshow goodness-of-fit test chi-square of 2.905 (6 degrees of freedom, P = 0.821), as well as good discrimination (C statistic = 0.768, CI 95% 0.661–0.876). In our pharmacogenetic study we have evaluated several clinical features available that seem to influence rituximab response. FLIPI is a prognostic index which predicts survival in FL patients lymphoma stage being part of this score whereas gender and β2M are not included although have traditionally been regarded as prognostic factors in FL 1. Our results indicate that FL patients carrying the FASL rs763110 T allele were associated with poorer response even after a multivariate analysis. Interestingly, the −844C allele showed increased FASL expression compared to the −844T 3. Moreover, some in vivo studies have reported that rituximab interferes in the FAS/FASL system, sensitizing B-NHL tumor cell lines to FASL 4. On note, FASL rs763110 T allele was previously associated with poorer response to a MAb in a different setting; i.e., anti-TNF IgG infliximab in Crohn's disease 3 thus highlighting the concept of a common mechanism of action for MAbs regardless their target molecule. This is the first study reporting that a polymorphism in an extrinsic apoptosis gene is associated with rituximab response in FL, along with gender, stage, and β2M. Patients with CC rs763110 polymorphism are better responders to rituximab therapy and this information might be taken into account when optimizing the use of anti-CD20 therapy. Rocío Gutiérrez-Cívicos,1 Ana María Hurtado,2 Daniel Torres-Moreno,3 José Javier Sanchez-Blanco,2 Ignacio Español,4 Luciano Consuegra-Sánchez,5 Elena Perez-Ceballos,2 María Dolores Gutiérrez-Meca,4 Andrés Jerez,2* Pablo Conesa-Zamora6* 1Pharmacy Department, Santa Lucía University Hospital, Cartagena, Spain 2Hematology and Medical Oncology Department, University Hospital Morales Meseguer. Centro Regional de Hemodonación. Universidad de Murcia. IMIB, Murcia, Spain 3Pathology Department, Santa Lucía University Hospital, Cartagena, Spain 4Hematology Department, Santa Lucía University Hospital, Cartagena, Spain 5Statistical Consulting Unit, Cardiology Department, Santa Lucía University Hospital, Cartagena, Spain 6Clinical Analysis Department, Santa Lucía University Hospital, Universidad Católica de Murcia (UCAM), Cartagena, Spain Additional Supporting Information may be found in the online version of this article. Supporting Information Table 1. Supporting Information Table 2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.