Absolute lymphocyte counts predicts response to ... - Nature

Letters to the Editor

2227 Table 1 DNA ligase IV A3V and T9I genotype distribution (wild type (CC), heterozygous (CT), homozygous variant (TT)) in pediatric ALL and controls DNA ligase IV genotype Wild type Polymorphism T9I heterozygous T9I homozygous A3V heterozygous/T9I heterozygous A3V heterozygous/T9I homozygous

ALL (n ¼ 107) 85 22 11 1 8 2

(79.4%) (20.6%) (10.3%) (0.9%) (7.5%) (1.9%)

Controls (n ¼ 104) 68 36 21 1 12 2

(65.4%) (34.6%) (20.2%) (1.0%) (11.5%) (1.9%)

Abbreviation: ALL, acute lymphoblastic leukemia.

gene are found in about 80% of infant leukemias, and NHEJ could be involved in these recombination events as well. Some features point to a potential involvement of NHEJ in the creation of the chromosomal translocation t(12;21), the most common genetic rearrangement in childhood ALL,1 resulting in the fusion protein ETV6/RUNX1 (TEL/AML1) (reviewed in Rassool2). Thus, it seems possible that a decrease of NHEJ activity could reduce the formation of potentially threatening frameshift mutations and translocations and, thereby, prevent the development of cancer. Whether or not the polymorphisms A3V and T9I alone and in heterozygous form could influence the LIGIV activity, still remains to be elucidated.

J Andreae1, R Varon2, K Sperling2 and K Seeger1 Department of Pediatric Oncology/Hematology, Charite´ Universita¨tsmedizin Berlin, Campus Virchow Klinikum, Berlin, Germany and 2 Institute of Human Genetics, Charite´ Universita¨tsmedizin Berlin, Campus Virchow Klinikum, Berlin, Germany E-mail: [email protected] 1

0.047 and 0.117, respectively. Overall, among the ALL patients only 20.6% carried at least one of the polymorphisms, while this rate was 34.6% within the control population. This difference is statistically significant (w2 ¼ 5.2266, Po0.025). The results are summarized in Table 1. No significant differences were observed concerning ALL immunophenotype, sex, age at diagnosis, survival rate, or risk of relapse. These results implicate that the two polymorphisms A3V and T9I do not increase the risk of ALL in children, but rather suggest that they could convey a protective effect. Roddam et al.6 reported a lower frequency of A3V and T9I in patients with multiple myeloma as compared to the control population and discussed their potential protective effects. However, in this study, no significant difference was found between ALL patients (n ¼ 70) and controls (n ¼ 220) regarding the frequencies of wild-type and variant LIGIV gene. Since multiple myeloma is not found in childhood and no age was provided for the ALL patients, it is reasonable to assume that this study population consisted of adults. In support of our study on childhood ALL, recent data indicate that T9I might reduce the risk for nonHodgkin’s lymphomas as well.8 The LIGIV isoform carrying the variants A3V and T9I has been shown to display a 2- to 3-fold reduced ligation activity.5 Considering the essential role of DNA repair for the maintenance of genomic stability and the important functions of NHEJ for the repair of DSB, a protective effect of polymorphisms potentially decreasing the NHEJ activity in a central component of the NHEJ pathway might be surprising. However, there is growing evidence that NHEJ could be error-prone and could in fact create chromosomal translocations frequently seen in leukemic cells. Chromosomal translocations involving the MLL

References 1 Armstrong SA, Look AT. Molecular genetics of acute lymphoblastic leukemia. J Clin Oncol 2005; 23: 6306–6315. 2 Rassool FV. DNA double strand breaks (DSB) and non-homologous end joining (NHEJ) pathways in human leukemia. Cancer Lett 2003; 193: 1–9. 3 Lieber MR, Ma Y, Pannicke U, Schwarz K. Mechanism and regulation of human non-homologous DNA end-joining. Nat Rev Mol Cell Biol 2003; 4: 712–720. 4 O’Driscoll M, Gennery AR, Seidel J, Concannon P, Jeggo PA. An overview of three new disorders associated with genetic instability: LIG4 syndrome, RS-SCID and ATR-Seckel syndrome. DNA Repair (Amst) 2004; 3: 1227–1235. 5 Girard PM, Kysela B, Harer CJ, Doherty AJ, Jeggo PA. Analysis of DNA ligase IV mutations found in LIG4 syndrome patients: the impact of two linked polymorphisms. Hum Mol Genet 2004; 13: 2369–2376. 6 Roddam PL, Rollinson S, O’Driscoll M, Jeggo PA, Jack A, Morgan GJ. Genetic variants of NHEJ DNA ligase IV can affect the risk of developing multiple myeloma, a tumour characterised by aberrant class switch recombination. J Med Genet 2002; 39: 900–905. 7 Goode EL, Dunning AM, Kuschel B, Healey CS, Day NE, Ponder BA et al. Effect of germ-line genetic variation on breast cancer survival in a population-based study. Cancer Res 2002; 62: 3052–3057. 8 Hill DA, Wang SS, Cerhan JR, Davis S, Cozen W, Severson RK et al. Risk of Non-Hodgkin lymphoma (NHL) in relation to germline variation in DNA repair and related genes. Blood 2006; 108: 3161–3167.

Absolute lymphocyte counts predicts response to chemotherapy and survival in diffuse large B-cell lymphoma

Leukemia (2007) 21, 2227–2230; doi:10.1038/sj.leu.2404780; published online 7 June 2007

The recovery of absolute lymphocyte count (ALC) after CHOP or rituximab plus CHOP (R-CHOP) chemotherapy has been determined as an independent prognostic factor of survival in

patients with diffuse large B-cell lymphoma (DLBCL).1 This finding suggests the importance of ALC recovery (that is, host immunity) in predicting the survival of DLBCL patients after chemotherapy.1,2 Furthermore, a recent study suggested that the ALC at the time of diagnosis, not at the time of recovery after chemotherapy, would predict the prognosis of 228 patients with follicular lymphoma (FL).3 FL patients with high ALCs at

Leukemia


diagnosis (X1.0 109/l) showed better overall survival compared to those with lower ALCs, and ALC was an independent prognostic factor in addition to the follicular lymphoma international prognostic index (FLIPI). However, to our best knowledge, the predictive value of the ALC at diagnosis for survival or response to chemotherapy has never been investigated in patients with DLBCL. The objective of the current study was to define the prognostic role of ALCs at diagnosis in predicting the response to chemotherapy and survival after treatment. A total of 223 patients with CD20-positive DLBCL who received CHOP or R-CHOP chemotherapy between 1997 and 2005 at six hospitals in Republic of Korea (Korean Cancer Center Hospital, Kyungpook National University Hospital, Chonnam National University Hwasun Hospital, Dongsan Medical Center, Pusan National University Hospital, and Yeungnam Medical Center) were included in this retrospective study. This study was approved by the institutional research board at the Kyungpook National University Hospital (Daegu, Korea). Clinical data were analyzed according to information available as of November 2005. Baseline patients’ characteristics are summarized in Supplementary Table 1. CHOP chemotherapy was given to 101 patients, and R-CHOP chemotherapy was given to 122 patients. No difference in baseline characteristics between the CHOP vs R-CHOP groups was noted (data not shown). Of the 223 patients, 12 patients received high-dose chemotherapy with autologous stem cell transplantation. Mean ALC (7s.e.) at the time of diagnosis was 1.7270.05 109/l. ALC at 25, 50 and 75 percentile was 1.13 109, 1.56 109 and 2.23 109/l, respectively. Using an ALC cut-off value of 1.0 109/l at diagnosis, patients were divided into higher and lower ALC groups. Forty-two patients

Probability of overall survival

a 1.0

(19%) had lower ALCs at diagnosis. Significant differences were noted between the higher and lower ALC groups in terms of stage (P ¼ 0.004), LDH level (Po0.001), extranodal involvement (P ¼ 0.011), performance status (Po0.001) and international prognostic index (IPI; Po0.001). These correlations suggest that ALCs may represent an additional prognostic parameter in patients with DLBCL. However, there was no association of ALCs with the use of rituximab (P ¼ 0.982), or age (P ¼ 0.539). Out of 215 patients evaluable for response to chemotherapy, the overall response rate (ORR) was 90% (194/215 patients) with a CR of 77% (165/215 patients) and a partial response (PR) rate of 13% (29/215 patients). A higher ORR was observed in the higher ALC group at diagnosis compared to the lower ALC group (94 vs 77%; P ¼ 0.005; Supplementary Table 1 and Supplementary Figure 1). The difference in ORR was significant in favor of the higher ALC group in patients receiving R-CHOP (ORR 95 vs 78%; P ¼ 0.019), but not in patients receiving CHOP chemotherapy (ORR 90 vs 77%, P ¼ 0.227). With a median follow-up duration of 376 days (range 12–5,176 days), 67 patients (30%) relapsed or progressed with 42 deaths (19%). The 1- and 2-year overall survival (OS) rates were 8773 and 7374%, while 1- and 2-year event-free survival (EFS) rates were 7373 and 5874%, respectively. A significant benefit in EFS was noted in the R-CHOP group compared with CHOP group (P ¼ 0.027). However, there was no statistically significant difference in OS between the R-CHOP vs CHOP groups (P ¼ 0.144), possibly due to a relatively short follow-up of the R-CHOP group. A survival benefit was noted in the higher ALC group compared to the lower ALC group with regard to both median duration of OS (3000 days vs 695days, Po0.001; Figure 1a) and EFS (1773 days vs 326days, Po0.001;

b High ALC group (n=181) 80±4 % at 2 years

0.8

p < 0.001 Overall patients

0.6

0.4

0.2

Low ALC group (n=42) 46±11 % at 2 years

0.0 0

2

4

6

8

10

12

Probability of event-free survival

2228

1.0

High ALC group (n=181) 64± 5 % at 2 years

0.8

0.6

0.4

0.2


0.0

14

0

2

Years from treatment

p < 0.001 R-CHOP group

0.6


0.4

6

8

10

12

14

d 1.0

High ALC group (n=103) 79±10 % at 2 years

0.8

4


0.2



c 1.0

p < 0.001 Overall patients

High ALC group (n=78) 77±5 % at 2 years

0.8

p = 0.001 CHOP group

0.6

0.4

0.2


0.0

0.0 0

1

2


3

0

2

4

6

8

10

12

14


Figure 1 Survival curve of diffuse large B-cell lymphoma patients according to the absolute lymphocyte counts at diagnosis in overall patients (a, b), in patients receiving R-CHOP regimen (c) and in patients receiving CHOP regimen (d). Leukemia


2229 Table 1

Multivariate analysis for the response and survival including event-free and overall survival

Overall response

Risk factor 9

Response rate (%) (95% CI)

OR (95% CI)

ALCs at diagnosis

X1.0 10 /l o1.0 109/l

93% (89–97%) 77% (64–91%)

1.000 2.717 (1.282–5.747)

EFS

Risk factor

2-years rate (%)

HR (95% CI)

ALCs at diagnosis

X1.0 109/l o1.0 109/l Score 0–2 Score 3–5 R-CHOP CHOP

64.375.0 31.879.7 71.975.4 32.577.1 67.877.7 52.575.5

1.000 2.148 (1.273–3.626) 1.000 2.932 (1.789–4.807) 1.000 1.768 (1.037–3.012)

2-years rate (%)

HR (95% CI)

79.774.3 55.079.3 86.373.9 52.278.1

1.000 2.863 (1.468–5.582) 1.000 3.861 (1.930–7.692)

IPI score Chemo regimen OS

Risk factor

ALCs at diagnosis

X1.0 109/l o1.0 109/l Score 0–2 Score 3–5

IPI score

P-value 0.009 P-value 0.004 o0.001 0.036 P-value 0.002 o0.001

Abbreviations: ALC, absolute lymphocyte count; CR, complete response; EFS, event-free survival; HR, hazard ratio; IPI, international prognostic score; OR, Odds ratio; OS, overall survival.

Figure 1b). The 2 years OS and EFS rates were 8074 and 6475% in the higher ALC group, and 46711 and 327 10% in the lower ALC group, respectively. In the R-CHOP group, the higher ALC group showed better survival than the lower ALC group in terms of OS (79710% vs 50717% at 2 years, Po0.001; Figure 1c) and EFS rate (66713% vs 23718% at 2 years, P ¼ 0.001; Supplementary Figure 2a). In the CHOP group, a survival difference was noted in favor of the higher ALC group compared to the lower ALC group in terms of OS (7775% vs 5876%, P ¼ 0.001; Figure 1d) and EFS (42715% vs 34711%, P ¼ 0.001; Supplementary Figure 2b). The results of the multivariate analysis are summarized in Table 1. To identify predictive factors for response to first line chemotherapy, a logistic regression analysis was conducted for 210 patients with following variables: stage, IPI score, age, performance status, LDH, extranodal involvement, use of rituximab and ALC at diagnosis (X1.0 vs o1.0 109/l). ALC at diagnosis was found to be an independent factor predicting overall response (P ¼ 0.009, Odds ratio 2.717, 95% CI (1.282–5.747)). For the multivariate survival analyses using Cox’s proportional hazard model to define prognostic factors for OS or EFS, the following variables were included into the final model: stage, IPI score, age, performance status, LDH level, extranodal involvement, use of rituximab and ALC at diagnosis (X1.0 vs o1.0 109/l). ALC at diagnosis was an independent prognostic factor for OS and EFS (Table 1). The lower ALC group had 31.6% of 2-year EFS, while the higher ALC group had 64.3% (P ¼ 0.004, HR 2.148, 95% CI. (1.273–3.626)). In addition, lower IPI at diagnosis and use of R-CHOP were found to be good prognostic factors with regard to EFS. The lower ALC group had 55.0% of 2-year OS, while the higher ALC group had 79.7% (P ¼ 0.002, HR 2.863, 95% CI (1.468– 5.582)). Lower IPI at diagnosis was found to be another good prognostic factor for OS, but the use of R-CHOP did not affect OS (Table 1). Despite correlation between ALC and other prognostic factors, multivariate analyses demonstrated that ALC is an independent prognostic factor for EFS and OS, suggesting the role of ALCs as another prognostic factor in patients with DLBCL.

The current study suggests that host immunocompetence may affect the response to chemotherapy and survival of patients with DLBCL after CHOP/R-CHOP therapy. Several prognostic factors have been reported for DLBCL, among which the IPI has been accepted as a single powerful prognostic factor for predicting the survival of patients with DLBCL.4 In addition, recent advances in immunotherapy emphasize the importance of host immunocompetence against the tumor. ALC recovery or ALC at diagnosis have been reported to be a surrogate marker of host immunocompetence. A cut-off value of 1.0 109/l was chosen based on a previous study in FL patients.3 Out of 223 patients, 42 patients (19%) were stratified into the lower ALC group. Based on the present result, the predictive power of ALC at diagnosis on EFS or OS was statistically significant irrespective of frontline chemotherapy regimen, including CHOP or R-CHOP therapy. However, the recovery of ALC after chemotherapy should be also considered as a predictor for successful treatment outcomes. A limitation of the current study includes a relatively short follow-up time, especially in the R-CHOP group. Another modifiable factor influencing patients’ prognosis is the use of a rituximab-containing regimen. Coiffier et al.5 reported that the group receiving R-CHOP instead of CHOP regimen had a survival benefit. The current study showed that the role of ALCs in predicting response to R-CHOP was more robust than response to CHOP chemotherapy, although the prognostic value of ALC was reproducible in both the R-CHOP and CHOP groups. One of the proposed mechanisms is antibody-dependent cellular cytotoxicity (ADCC) in addition to bcl-2-mediated chemoresistance.6,7 Increasing evidence supports the role of natural killer (NK) cells as important mediators of ADCC after binding of rituximab to CD20-positive lymphoma cells. Manches et al.6 reported that rituximab can induce phagocytosis of lymphoma cells by tissue-scavengering cells such as macrophages or NK cells, important mediators of ADCC. Accordingly, host immunocompetence, reflected by ALC at diagnosis, would predict the response to chemotherapy (especially R-CHOP) and prognosis of patients with DLBCL. Further study to determine lymphocyte subsets responsible for the superior response and outcomes will provide us better understanding of this issue. Leukemia


2230 In conclusion, the present study suggests that the ALC at diagnosis (X1.0 109/l) is predictive of response to chemotherapy (especially R-CHOP) and can be a good prognostic factor for DLBCL patients. Host immunocompetence at the time of diagnosis will be an important parameter of prognosis in patients with DLBCL.

DH Kim1,2, JH Baek1,3, YS Chae1, Y-K Kim4, HJ Kim4, YH Park2,5, HS Song6, JS Chung7, MS Hyun8 and SK Sohn1 1 Department of Hematology/Oncology, Kyungpook National University Hospital, Daegu, Korea; 2 Department of Hematology/Oncology, Samsung Medical Center, Sungkyunkwan University Department of School of Medicine, Seoul, Korea; 3 Department of Hematology/Oncology, Ulsan University Hospital, University of Ulsan, Ulsan, Korea; 4 Department of Hematology/Oncology, Chonnam National University Hwasun Hospital, Jeollanamdo, Korea; 5 Department of Hematology/Oncology, Korean Cancer Center Hospital, Seoul, Korea; 6 Department of Hematology/Oncology, Keimyung University, Dongsan Medical Center, Daegu, Korea; 7 Department of Hematology/Oncology, Pusan National University Hospital, Busan, Korea and 8 Department of Hematology/Oncology, Yeungnam University, Yeungnam Medical Center, Daegu, Korea E-mail: [email protected]

References 1 Porrata LF, Ristow KM, Markovic SN, Persky D, Habermann TM. Lymphocyte count persistence and early recovery predicts superior survival and is independent of the international prognostic index in patients treated with CHOP chemotherapy for diffuse large B cell lymphoma. Blood 2004; 104: 3252. 2 Porrata LF, Ristow KM, Geyer SM, Markovic SN, Persky DO, Colgan JP et al. Absolute lymphocyte count recovery predicts superior survival and is independent of the international prognostic index in patients treated with CHOP or R-CHOP chemotherapy for diffuse large B cell lymphoma. Blood 2005; 106: 931. 3 Siddiqui M, Ristow K, Markovic SN, Witzig TE, Habermann TM, Colgan JP et al. Absolute lymphocyte count predicts overall survival in follicular lymphomas. Br J Haematol 2006; 134: 596–601. 4 A predictive model for aggressive non-Hodgkin’s lymphoma. The international non-Hodgkin’s lymphoma prognostic factors project. N Engl J Med 1993; 329: 987–994. 5 Coiffier B, Lepage E, Briere J, Herbrecht R, Tilly H, Bouabdallah R et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346: 235–242. 6 Manches O, Lui G, Chaperot L, Gressin R, Molens JP, Jacob MC et al. In vitro mechanisms of action of rituximab on primary non-Hodgkin’s lymphomas. Blood 2003; 101: 949–954. 7 Cartron G, Watier H, Golay J, Solal-Celigny P. From the bench to the bedside: ways to improve rituximab efficacy. Blood 2004; 104: 2635–2642.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Adult ALL: treatment outcome and prognostic factors in an Indian population using a modified German ALL (GMALL) protocol

Leukemia (2007) 21, 2230–2233; doi:10.1038/sj.leu.2404785; published online 7 June 2007

Treatment outcome in adult acute lymphoblastic leukemia (ALL) has shown little improvement over the years with complete remission (CR) rates of 80–85% and leukemia-free survival of 30– 40%.1–3 The results of treatment have significantly improved in Tcell and mature B-cell ALL, without much change in B-precursor ALL.4 Attempts to improve outcome include high-dose intensification, stem cell transplantation and risk-adapted therapy based on identified prognostic markers or minimal residual disease. However, in developing countries where state funding for expensive health-care interventions is lacking, it is important that treatment modifications should be cost effective.5 This is a retrospective analysis of the outcome of 210 adult patients with ALL (X15 years) treated with a modified German ALL (GMALL) protocol from 1994 to 2003. The aim of this study was to identify biologic and clinical prognostic markers for adult ALL in the Indian population and compare this with Caucasian data. The diagnosis of ALL was based on morphological evaluation of May Grunwald Giemsa-stained smears of bone marrow and confirmed by cytochemistry and immunophenotypic analysis. Central nervous system (CNS) involvement was diagnosed if more than 5 cells per ml were present in a nontraumatic cerebro spinal fluid (CSF) at diagnosis and if lymphoblasts were identified on cytospin. Karyotypic analysis was not carried out systematically in this cohort of patients. Leukemia

After preinduction with steroids for 1 week, phase I induction was given over 4 weeks with vincristine (VCR), daunorubicin (DNR), L-asparaginase (ASP) and prednisolone (PDN). Bone marrow examination was done after completion of the phase I induction to assess remission. Phase II induction consisted of cyclophosphamide (Cyclo) and four courses of cytosine (Ara C) combined with CNS-directed therapy using intrathecal methotrexate (MTx) and prophylactic cranial radiotherapy. Consolidation consisted of two 5-day courses of Ara C (75 mg/m2) and etoposide (50 mg/m2) along with a single dose of intrathecal MTx (12.5 mg). This was followed by re-induction with VCR, DNR, dexamethasone (Dexa), Cyclo and Ara C over a period of 6 weeks. The final maintenance consisted of daily mercaptopurine (6-MP) and weekly oral MTx for 24 months. This maintenance was intensified with monthly pulses of VCR 2 mg, Dexa 10 mg/m2 for 5 days and intrathecal MTx 12.5 mg once in 3 months (Figure 1). The differences with the GMALL protocol are: six thioguanine omitted during phase II induction and consolidation and addition of VCR þ Dexa pulses every 4 weeks þ intrathecal MTx once in 3 months during maintenance for a total period of 24 months. The management of febrile neutropenic episodes was carried out according to standard guidelines. Good PDN response (GPR) was defined as o1000 blasts/ml in the peripheral blood on day 8 of starting treatment. At the end of phase I induction, patients were considered to be in CR if the peripheral blood count was normal with no blasts in the peripheral smear or CSF and no lymphoblasts in marrow