PROGNOSTIC FACTORS IN PRIMARY MEDIASTINAL LARGE B-CELL LYMPHOMA (PMLBCL) AFTER RITUXIMAB-CHOP (R-CHOP) WITH OR WITHOUT RADIOTHERAPY (RT): MATURE RESULTS OF A COOPERATIVE RETROSPECTIVE STUDY
(Abstract release date: 05/19/16)
EHA Library. Vassilakopoulos T. 06/09/16; 132499; E950
Assoc. Prof. Theodoros Vassilakopoulos
Contributions
Contributions
Abstract
Abstract: E950
Type: Eposter Presentation
Background
Prognostic factors (PFs) have not been extensively studied in PMLBCL and prognostic models specifically applicable to this entity have not been developed, mainly due to its rarity. R-CHOP provides very good results in PMLBCL, minimizing failure rates. High IPI and serous effusions emerged as adverse prognostic factors in 181 Japanese patients treated with RCHOP±RT (selected among broader population with heterogenous treatment), while serous effusions, B-symptoms and age were identified in 96 RCHOP±RT-treated patients in a 2012 abstract from Vancouver. Given that more intensive chemotherapy (R-da-EPOCH) might be better than R-CHOP, the applicability of various PFs needs to be urgently evaluated in the Rituximab era in order to define subgroups of patients at high risk for treatment failure and death.
Aims
The identification of PFs for the outcome of patients with PMLBCL treated with RCHOP± RT.
Methods
213 patients with PMLBCL were treated in a multicenter setting with RCHOP±RT (usually 6-8 cycles). The following potential prognostic factors were evaluated: Age (median 31; range 17-82; >60 years only 4%), gender (female 64%), B-symptoms, stage III/IV, infradiaphragmatic disease, extranodal involvement (either stage IV or stage E), pleuritis, pericarditis, performance status (PS) ≥2, LDH levels, anemia, leukocytosis ≥10x109/L, ESR ≥30 mm/h, albumin <4g/dL, bulky disease (≥10 cm), age-adjusted IPI (aaIPI; ≥2 in 21%). A modified version of aaIPI was also analyzed, attributing 1 point to either stage III/IV or E-disease instead of stage III/IV only (aaIPI-mod).
Results
With 52 failures recorded (51 within 17 months from diagnosis), the 3-year freedom from progression (FFP) was 75%. With 24 deaths recorded (excluding 2 unrelated deaths), the 5-year overall survival (OS) was 87%. aaIPI≥2 identified a minority of patients (21%) with a 5-year FFP of 64% vs. 79% for those with aaIPI 0-1 (p=0.04) and 5-year OS of 76% vs. 91% (p=0.0095). The aaIPI-mod was more effective in predicting the outcome and indentified a larger poor prognostic group (41% of total): The 5-year FFP was 61% vs. 86% for those with aaIPI-mod 0-1 (p<0.0001), while 5-year OS was 75% vs. 97% (p<0.0001). Many of the examined variables had a significant (p<0.05) or borderline (p<0.15) association with both FFP and OS (extranodal disease, LDH, PS, abdominal disease, bulky disease, serous effusions, anemia). In multivariate analysis of FFP, extranodal involvement and bulky disease were independent PFs (p=0.006 and p=0.04). None, 1 or 2 of these factors were present in 29%, 42% and 30% of the patients. FFP at 5 years was effectively predicted being 88%, 79% and 59% for patients with 0, 1 or 2 PFs respectively, while 5-year disease specific survival was 100%, 91% and 72%.
Conclusion
In the largest patient series reported so far, RCHOP±RT provided satisfactory results in PMLBCL with long-term FFS of 75% and excellent OS of 87%. The aaIPI was moderately predictive of the outcome but a modified version performed better. Either the aaIPI-mod or the combination of extranodal involvement and bulky disease defined a subgroup comprising ~40% and ~30% of patients respectively with a ~40% risk of failure and ≥25% risk of death, who might be suitable for trials of treatment intensification.
Session topic: E-poster
Type: Eposter Presentation
Background
Prognostic factors (PFs) have not been extensively studied in PMLBCL and prognostic models specifically applicable to this entity have not been developed, mainly due to its rarity. R-CHOP provides very good results in PMLBCL, minimizing failure rates. High IPI and serous effusions emerged as adverse prognostic factors in 181 Japanese patients treated with RCHOP±RT (selected among broader population with heterogenous treatment), while serous effusions, B-symptoms and age were identified in 96 RCHOP±RT-treated patients in a 2012 abstract from Vancouver. Given that more intensive chemotherapy (R-da-EPOCH) might be better than R-CHOP, the applicability of various PFs needs to be urgently evaluated in the Rituximab era in order to define subgroups of patients at high risk for treatment failure and death.
Aims
The identification of PFs for the outcome of patients with PMLBCL treated with RCHOP± RT.
Methods
213 patients with PMLBCL were treated in a multicenter setting with RCHOP±RT (usually 6-8 cycles). The following potential prognostic factors were evaluated: Age (median 31; range 17-82; >60 years only 4%), gender (female 64%), B-symptoms, stage III/IV, infradiaphragmatic disease, extranodal involvement (either stage IV or stage E), pleuritis, pericarditis, performance status (PS) ≥2, LDH levels, anemia, leukocytosis ≥10x109/L, ESR ≥30 mm/h, albumin <4g/dL, bulky disease (≥10 cm), age-adjusted IPI (aaIPI; ≥2 in 21%). A modified version of aaIPI was also analyzed, attributing 1 point to either stage III/IV or E-disease instead of stage III/IV only (aaIPI-mod).
Results
With 52 failures recorded (51 within 17 months from diagnosis), the 3-year freedom from progression (FFP) was 75%. With 24 deaths recorded (excluding 2 unrelated deaths), the 5-year overall survival (OS) was 87%. aaIPI≥2 identified a minority of patients (21%) with a 5-year FFP of 64% vs. 79% for those with aaIPI 0-1 (p=0.04) and 5-year OS of 76% vs. 91% (p=0.0095). The aaIPI-mod was more effective in predicting the outcome and indentified a larger poor prognostic group (41% of total): The 5-year FFP was 61% vs. 86% for those with aaIPI-mod 0-1 (p<0.0001), while 5-year OS was 75% vs. 97% (p<0.0001). Many of the examined variables had a significant (p<0.05) or borderline (p<0.15) association with both FFP and OS (extranodal disease, LDH, PS, abdominal disease, bulky disease, serous effusions, anemia). In multivariate analysis of FFP, extranodal involvement and bulky disease were independent PFs (p=0.006 and p=0.04). None, 1 or 2 of these factors were present in 29%, 42% and 30% of the patients. FFP at 5 years was effectively predicted being 88%, 79% and 59% for patients with 0, 1 or 2 PFs respectively, while 5-year disease specific survival was 100%, 91% and 72%.
Conclusion
In the largest patient series reported so far, RCHOP±RT provided satisfactory results in PMLBCL with long-term FFS of 75% and excellent OS of 87%. The aaIPI was moderately predictive of the outcome but a modified version performed better. Either the aaIPI-mod or the combination of extranodal involvement and bulky disease defined a subgroup comprising ~40% and ~30% of patients respectively with a ~40% risk of failure and ≥25% risk of death, who might be suitable for trials of treatment intensification.
Session topic: E-poster
Abstract: E950
Type: Eposter Presentation
Background
Prognostic factors (PFs) have not been extensively studied in PMLBCL and prognostic models specifically applicable to this entity have not been developed, mainly due to its rarity. R-CHOP provides very good results in PMLBCL, minimizing failure rates. High IPI and serous effusions emerged as adverse prognostic factors in 181 Japanese patients treated with RCHOP±RT (selected among broader population with heterogenous treatment), while serous effusions, B-symptoms and age were identified in 96 RCHOP±RT-treated patients in a 2012 abstract from Vancouver. Given that more intensive chemotherapy (R-da-EPOCH) might be better than R-CHOP, the applicability of various PFs needs to be urgently evaluated in the Rituximab era in order to define subgroups of patients at high risk for treatment failure and death.
Aims
The identification of PFs for the outcome of patients with PMLBCL treated with RCHOP± RT.
Methods
213 patients with PMLBCL were treated in a multicenter setting with RCHOP±RT (usually 6-8 cycles). The following potential prognostic factors were evaluated: Age (median 31; range 17-82; >60 years only 4%), gender (female 64%), B-symptoms, stage III/IV, infradiaphragmatic disease, extranodal involvement (either stage IV or stage E), pleuritis, pericarditis, performance status (PS) ≥2, LDH levels, anemia, leukocytosis ≥10x109/L, ESR ≥30 mm/h, albumin <4g/dL, bulky disease (≥10 cm), age-adjusted IPI (aaIPI; ≥2 in 21%). A modified version of aaIPI was also analyzed, attributing 1 point to either stage III/IV or E-disease instead of stage III/IV only (aaIPI-mod).
Results
With 52 failures recorded (51 within 17 months from diagnosis), the 3-year freedom from progression (FFP) was 75%. With 24 deaths recorded (excluding 2 unrelated deaths), the 5-year overall survival (OS) was 87%. aaIPI≥2 identified a minority of patients (21%) with a 5-year FFP of 64% vs. 79% for those with aaIPI 0-1 (p=0.04) and 5-year OS of 76% vs. 91% (p=0.0095). The aaIPI-mod was more effective in predicting the outcome and indentified a larger poor prognostic group (41% of total): The 5-year FFP was 61% vs. 86% for those with aaIPI-mod 0-1 (p<0.0001), while 5-year OS was 75% vs. 97% (p<0.0001). Many of the examined variables had a significant (p<0.05) or borderline (p<0.15) association with both FFP and OS (extranodal disease, LDH, PS, abdominal disease, bulky disease, serous effusions, anemia). In multivariate analysis of FFP, extranodal involvement and bulky disease were independent PFs (p=0.006 and p=0.04). None, 1 or 2 of these factors were present in 29%, 42% and 30% of the patients. FFP at 5 years was effectively predicted being 88%, 79% and 59% for patients with 0, 1 or 2 PFs respectively, while 5-year disease specific survival was 100%, 91% and 72%.
Conclusion
In the largest patient series reported so far, RCHOP±RT provided satisfactory results in PMLBCL with long-term FFS of 75% and excellent OS of 87%. The aaIPI was moderately predictive of the outcome but a modified version performed better. Either the aaIPI-mod or the combination of extranodal involvement and bulky disease defined a subgroup comprising ~40% and ~30% of patients respectively with a ~40% risk of failure and ≥25% risk of death, who might be suitable for trials of treatment intensification.
Session topic: E-poster
Type: Eposter Presentation
Background
Prognostic factors (PFs) have not been extensively studied in PMLBCL and prognostic models specifically applicable to this entity have not been developed, mainly due to its rarity. R-CHOP provides very good results in PMLBCL, minimizing failure rates. High IPI and serous effusions emerged as adverse prognostic factors in 181 Japanese patients treated with RCHOP±RT (selected among broader population with heterogenous treatment), while serous effusions, B-symptoms and age were identified in 96 RCHOP±RT-treated patients in a 2012 abstract from Vancouver. Given that more intensive chemotherapy (R-da-EPOCH) might be better than R-CHOP, the applicability of various PFs needs to be urgently evaluated in the Rituximab era in order to define subgroups of patients at high risk for treatment failure and death.
Aims
The identification of PFs for the outcome of patients with PMLBCL treated with RCHOP± RT.
Methods
213 patients with PMLBCL were treated in a multicenter setting with RCHOP±RT (usually 6-8 cycles). The following potential prognostic factors were evaluated: Age (median 31; range 17-82; >60 years only 4%), gender (female 64%), B-symptoms, stage III/IV, infradiaphragmatic disease, extranodal involvement (either stage IV or stage E), pleuritis, pericarditis, performance status (PS) ≥2, LDH levels, anemia, leukocytosis ≥10x109/L, ESR ≥30 mm/h, albumin <4g/dL, bulky disease (≥10 cm), age-adjusted IPI (aaIPI; ≥2 in 21%). A modified version of aaIPI was also analyzed, attributing 1 point to either stage III/IV or E-disease instead of stage III/IV only (aaIPI-mod).
Results
With 52 failures recorded (51 within 17 months from diagnosis), the 3-year freedom from progression (FFP) was 75%. With 24 deaths recorded (excluding 2 unrelated deaths), the 5-year overall survival (OS) was 87%. aaIPI≥2 identified a minority of patients (21%) with a 5-year FFP of 64% vs. 79% for those with aaIPI 0-1 (p=0.04) and 5-year OS of 76% vs. 91% (p=0.0095). The aaIPI-mod was more effective in predicting the outcome and indentified a larger poor prognostic group (41% of total): The 5-year FFP was 61% vs. 86% for those with aaIPI-mod 0-1 (p<0.0001), while 5-year OS was 75% vs. 97% (p<0.0001). Many of the examined variables had a significant (p<0.05) or borderline (p<0.15) association with both FFP and OS (extranodal disease, LDH, PS, abdominal disease, bulky disease, serous effusions, anemia). In multivariate analysis of FFP, extranodal involvement and bulky disease were independent PFs (p=0.006 and p=0.04). None, 1 or 2 of these factors were present in 29%, 42% and 30% of the patients. FFP at 5 years was effectively predicted being 88%, 79% and 59% for patients with 0, 1 or 2 PFs respectively, while 5-year disease specific survival was 100%, 91% and 72%.
Conclusion
In the largest patient series reported so far, RCHOP±RT provided satisfactory results in PMLBCL with long-term FFS of 75% and excellent OS of 87%. The aaIPI was moderately predictive of the outcome but a modified version performed better. Either the aaIPI-mod or the combination of extranodal involvement and bulky disease defined a subgroup comprising ~40% and ~30% of patients respectively with a ~40% risk of failure and ≥25% risk of death, who might be suitable for trials of treatment intensification.
Session topic: E-poster
{{ help_message }}
{{filter}}