LRF CLL4 Research Articles

Quality of Life in the LRF CLL4 Trial.

The UK CLL4 trial randomised 777 previously untreated patients between 1999–2004 to chlorambucil (Chl) or fludarabine +/− cyclophosphamide (F v. FC). Other trial endpoints are reported elsewhere in this meeting (Catovsky et al). The current study seeks to determine whether there is any difference in the quality of life (QoL) of patients according to treatment arm; and whether QoL correlates with baseline patient characteristics and/or with treatment outcomes. All patients were asked to complete the validated EORTC QLQ-C30 (version 2.0) at start of therapy (baseline), at 3, 6 and 12 months, and annually thereafter. Scores (0–100) within each of 15 different measures of QoL (5 functional scales, 3 symptom scales, 1 global health/QoL scale, and 6 single items) were calculated at each point in time (currently up to 1 year), and analysed according to the EORTC QLQ-C30 Scoring Manual. Statistical significance was determined by the use of ANOVA. Patient compliance was high: 599 questionnaires were completed at baseline, 557 at 3 months, 559 at 6 months and 470 at 12 months, respectively 77%, 74%, 77% and 78% of those who had so far reached the time-period and remained alive. Analysis of baseline and 12 month QoL assessment by treatment group and response is presented. QOL results were the same for each treatment group at baseline and at 12 months (Table 1). Quality of response to treatment correlated with QoL at 12 months (Table 2). With treatment approaches having potentially different risk and side effect profiles, and perhaps limited impact on overall survival, QOL forms an important component of treatment outcome assessment. Further analysis of QoL parameters (in particular their relationship to prognostic factors and further follow-up) will be forthcoming within the next year.Table 1:Mean QoL Scores by Treatment Group*Functioning ScalesPhysicalRoleEmotionalCognitiveSocialGlobal HealthNo.B 12MB 12MB 12MB 12MB 12MB 12MChl17581 7874 7575 7883 8279 7766 69F9282 8272 7480 8182 8280 7964 68FC9480 7777 7378 8185 8282 7666 68361Abbreviations: B = baseline; 12M = 12 months.Table 2:Mean QoL Scores by Response to Treatment*Functioning ScalesPhysicalRoleEmotionalCognitiveSocialGlobal HealthNo.B 12MB 12MB 12MB 12MB 12MB 12MCR/NPR15984 8476 8078 8486 8582 8367 73PR12579 7673 7378 7882 7980 7763 68NR/PD5677 7471 6174 7481 7974 6565 60340p=0.02p=0.03p=0.002Abbreviations: B = baseline; 12M = 12 months; CR/NPR = complete/nodular partial response; PR = partial response; NR/PD = no response/disease progression. P-values compare response groups with respect to the change in scores from baseline to 12 months.(* Only those patients are shown in the Tables for whom data is available at both time-points.)

Prognostic Factors in the UK LRF CLL4 Trial.

The CLL4 trial randomised 777 previously untreated patients between Chlorambucil (Chlor), Fludarabine (Fluda) and Fludarabine with Cyclophosphamide (FC). The median follow-up is 21 months. VH gene mutation status using a 98% cut off, CD38 expression using a 7% cut off, ZAP70 expression (Orchard et al Lancet 2004) and interphase FISH were measured on samples taken at randomisation. In an interim analysis 253/397 patients had unmutated VH genes. 32 patients utilised the VH3-21 gene, of whom 27 showed restricted HCDR3 usage. 343/535 were CD38 positive and 131/261 were ZAP70 positive. 10/24 VH 3-21 cases were ZAP 70 positive, of whom 7 were unmutated. There was no significant association between VH genes and age, sex or a positive DAT. There were significant correlations between VH gene mutations and CD38, ZAP70 and FISH results.Association of VH genes with CD38, ZAP70 and FISH results.CD38+ZAP 70+p53 lossDel 11qDel 6q21Del 13qTrisomy 12VH Unmutated119 (58%)110 (72%)21 (9%)48 (22%)18 (10%)109 (45%)50 (21%)VH Mutated30 (25%)17 (16%)3 (2%)13 (11%)6 (6%)97 (73%)13 (10%)p=<0.0001<0.00010.02<0.0090.2<0.00010.008% = % of all cases with known VH mutation statusLogistic regression including age, stage, gender, VH mutations and genetic factors shows that stage (p=0.002), VH genes (p=0.05), p53 loss (p=0.02) and del 11q (p=0.01) affect response. Response rates by VH gene mutation status and treatment arm are given in table 2.Response rates according to VH mutation status and treatment armChlor UnmutChlor MutFluda UnmutFluda MutF/C UnmutF/C MutCR/NPR24 (25%)23 (42%)25 (43%)18 (55%)37 (58%)23 (70%)PR43 (45%)19 (34%)21 (36%)10 (30%)21 (33%)10 (30%)NR/PD29 (30%)13 (24%)12 (21%)5 (15%)6 (9%)0 (0%)p (trend)0.070.30.1%=% of all unmutated or mutated cases receiving a particular treatmentAnalysis of variance shows that both VH genes (p=0.01) and treatment (p<0.0001) have an effect on response with no significant interaction. Among 104 patients with unmutated VH genes and both ZAP 70 expression and response data available, there was no significant difference in response rates between ZAP positive and negative cases. In univariate analysis p53 loss (p<0.00005), unmutated VH genes (p=0.003), CD38 positivity (p=0.06) but not ZAP70 positivity correlate with poor overall survival. In multivariate Cox regression analyses, age (p<0.0001) stage (p=0.01) and p53 loss (p<0.0001) are significant for survival. When patients with p53 loss are excluded unmutated VH genes (p=0.007) but not CD38 or ZAP70 positivity are significant. In conclusion p53 loss identifies a small group of patients with both poor response to treatment and short survival. In the remaining cases, unmutated VH genes are associated with a relatively poor response independent of treatment arm, and with a shorter survival.

Comparison of Oral and Intravenous Fludarabine in the LRF CLL4 Trial.

The LRF CLL4 trial opened in 2/99 and closed in 10/04. 783 previously untreated patients with CLL requiring therapy by conventional criteria were entered and 6 were excluded. Patients were randomized between chlorambucil (10mg/m2 x 7 days; n=387) and fludarabine alone (FDR; n=194) or combined with cyclophosphamide (FC; n=196). A full report of the responses and outcomes of the trial will be presented elsewhere in the Meeting (Catovsky et al.). FDR and FC were initially given intravenously (IV; FDR 25mg/m2 x 5 days or FDR 25mg/m2 plus cyclophosphamide 250mg/m2 for 3 days) until 2001 when oral FDR became available and the protocol was modified to allow it's use. The oral dose of fludarabine (FDR 40mg/m2 x 5 days or FDR 24mg/m2 plus cyclophosphamide 150mg/m2 for 5 days) was considered equivalent to the IV dose as previous studies indicated that the bioavailability of oral fludarabine was approximately 55% of the IV route. Dosing in the chlorambucil arm of the trial remained unchanged. The chemotherapy dose was reduced if there was a 2 week delay in therapy due to neutropenia or thrombocytopenia and the fludarabine dose was reduced if the creatinine clearance was between 30 to 60ml/min. 223 patients received oral FDR (107 FDR alone and 116 FC) and 106 patients received IV FDR (51 FDR alone and 55 FC). 6 FDR and 15 FC patients were not assessable for response. Among the remainder, responses were better with IV fludarabine compared to oral therapy (FDR: CR/NPR = 54% v 41%, NR = 8% v 26%, p=0.02; FC: CR/NR = 73% v 59%, NR = 2% v 10%, p=0.04). There was no difference between the oral and IV patients in age, gender or stage, but as expected the proportion receiving oral therapy increased over the years (2001 75%, 2002 85%, 2003 89%, 2004 91%; p for trend=0.008) and patients entered after 2001 were older (mean 64.4 versus 62.5; p=0.02). However the response rate also decreased by year of entry in the chlorambucil arm (1999/2000 vs 2001/02 vs 2003/04: CR/NPR = 33% vs 30% vs 20%; NR/PD = 17% vs 28% vs 38%, p=0.004) with no significant difference over time when the responses for the two fludarabine-containing regimes, including oral and IV, were combined (1999/2000 vs 2001/02 vs 2003/04: CR/NPR = 57% vs 51% vs 51%; NR/PD = 6% vs 13% vs 16%, p=0.16). In multivariate Cox regression analysis of progression free survival (PFS) for patients entered from 2001, including treatment (FDR or FC), age, gender, stage, year of entry and IV or oral FDR as possible variables in the model, treatment (p=0.0003) and IV/oral (Hazard ratio (HR) = 1.9; p=0.05) were significant. Age was the only significant variable in analysis of survival (HR = 1.05; p=0.0003). There was no significant difference in the main toxicities (neutropenia, thrombocytopenia, anaemia, nausea and vomiting, alopecia, diarrhoea) for oral or IV fludarabine. In summary, analysis of the data suggests that the observed difference in response rates between intravenous and oral fludarabine in the LRF CLL4 trial is probably not due to the route of administration of fludarabine but is more likely to be explained because older patients with a poorer prognosis were entered later in the study when all patients were guaranteed to receive oral therapy. Although a randomised controlled trial would be required to formally prove that oral fludarabine is not inferior to intravenous fludarabine our data are suggestive that there are unlikely to be significant differences in response rates between routes of administration of fludarabine.

ZAP-70 Expression in Chronic Lymphocytic Leukemia: Correlation with Clinical and Biological Features.

One of the most intriguing features of chronic lymphocytic leukemia (CLL) is its clinical heterogeneity. The mutational status of immunoglobulin heavy chain variable region (IgVH) genes is one of the most powerful predictors of overall survival (OS) and progression-free survival (PFS). The expression of the tyrosine kinase ZAP-70 in CLL cells has been proposed as a surrogate marker for the mutational status of IgVH genes. Recent reports suggest that ZAP-70 over-expression is an independent prognostic factor for shorter OS and PFS. There is little information about correlation between ZAP-70 expression and other clinical and biological features in CLL. The aims of this study are 1) to evaluate prospectively ZAP-70 expression in a series of CLL patients; 2) to correlate this with clinical stage, lymphocyte morphology, CD38 expression and chromosomal abnormalities detected by FISH and 3) to analyze the independent prognostic value of ZAP-70 in predicting time to first treatment (treatment free interval, TFI). ZAP-70 expression was analyzed using four-colour flow cytometry (as described by Crespo et al, NEJM 2003; 348:1764–75). 201 previously untreated CLL patients were evaluated at diagnosis (32%) or during the disease course. Amongst them, 143 patients were entered in the LRF CLL4 trial. Clinical stage at the time of the study was A in 92 patients (A progressive in 46), B in 63 and C in 46. 28% were ZAP-70 positive (+) (cut-off ≥ 20% of CD5/CD19+ cells). ZAP-70 positivity was associated with prevalence of stage B/C at diagnosis (p=0.004), CLL with >10% prolymphocytes (CLL/PL) or atypical morphology (p<0.001), CD38 positivity (regardless of the cut-off point) (p<0.001), trisomy12 (p=0.001), del(6)(q21) (p=0.05) or no detectable abnormalities (0.002). ZAP-70 negative (−) cases presented more often with del(13)(q14) (p<0.001). There were no significant differences in the frequency of del(11)(q23) or delp53 between ZAP-70(+) and ZAP-70(−) cases. Median time from diagnosis to first treatment (treatment free interval, TFI) was 17.7 months for ZAP-70(+) and 44.6 months for ZAP-70(−) cases (p<0.001). The presence of the following parameters was associated with short TFI in univariate analysis: male sex (p=0.02), atypical morphology (p=0.02), clinical stage B/C (p<0.001), ZAP-70(+) (p<0.001), CD38+ (p<0.001), absence of del(13)(q14) (p<0.001), presence of delp53 (p=0.05). Multivariate analysis is being performed. In conclusion, we document a significant correlation between ZAP-70 over-expression in CLL and other adverse prognostic factors (such as CD38 and advanced stage), provide new data concerning its association with cytogenetic abnormalities detected by FISH and demonstrate the prognostic value of ZAP-70 expression.

Incidence of Hemolytic Anemia after Chemotherapy in the CLL4 Trial. A Possible Protective Role for Fludarabine Plus Cyclophosphamide.

In 1966, Lewis, Schwartz and Dameshek described in a landmark paper the triggering of autoimmune complications in CLL and related conditions by alkylating agents and radiotherapy (Lewis et al, Clin Exp Immunol 1, 3–11). This has been confirmed by many since and it is well known that autoimmune hemolytic anemia (AHA) is also a feature of untreated CLL but it is more common in treated cases. Review of data from MRC CLL trials in the last 20 years shows an incidence of AHA of 8.6% (1273 patients), which rises to 11% (716 patients) after therapy based largely on alkylating agents. The advent of nucleoside analogues and, in particular, fludarabine (FDR), showed that these agents could also trigger AHA. Because of the severity of this complication in some of the reported cases, there is a perception that FDR may induce AHA more often than other agents. The LRF CLL4 trial, launched in 1999, and still open, gave us the opportunity to document AHA in newly diagnosed patients with Binet stages A progressive, B and C randomised to: 1) chlorambucil 10mg/m2 x 7 days and 2) FDR alone x 5 days IV or oral or FDR plus cyclophosphamide x 3 days IV or x 5 days orally, all every 4 weeks. Information was available from follow-up forms after completion of chemotherapy on 470 randomised patients. AHA was reported in 23/218 (10.5%) after chlorambucil, 9/123 (7.3%) after FDR and 3/129 (2.3%) after FDR plus cyclophosphamide (Chi-square 7.95; p<0.02). The incidence in both FDR regimens combined was 4.8% (p<0.01 for the comparison with chlorambucil). Information on the status of the direct antiglobulin test (DAT) at entry was available on 379 patients; the overall incidence of DAT positivity was 14%. AHA developed in 20/326 DAT negative (6%) and 12/53 DAT positive (22.6%) patients (Chi-square 10.55; p<0.005). In both groups, AHA was more frequent after chlorambucil particularly in the DAT positive cases (p<0.02): 8% in DAT negative and 39% in DAT positive compared with FDR 7% in DAT negative and 18% in DAT positive cases and FDR plus cyclophosphamide 2% and 5.8%, respectively. Furthermore, of 28 patients who were DAT positive at entry we observed a change to DAT negativity in 7/9 with FDR plus cyclophosphamide, 6/13 with chlorambucil and 1/6 with FDR after completion of therapy (Chi-square 5.5 p = 0.063)We conclude that: 1) chlorambucil is more likely than FDR +/− cyclophosphamide to trigger AHA in CLL; 2) Although AHA could develop regardless of DAT status its incidence is higher in DAT positive cases; 3) The incidence of AHA is significantly lower in patients receiving FDR plus cyclophosphamide, suggesting a protective role for this combination which is supported by a higher conversion rate from DAT positive to DAT negative after therapy. A lower incidence of AHA was reported with FDR plus cyclophosphamide compared to FDR alone by the German CLL Study Group (Eichorst et al, Abstract 243, ASH 2003). We still need to establish whether the severity of AHA after FDR alone is greater than with alkylating agents, as 2 patients in our trial died of AHA after FDR.

Gender as a Prognostic Factor in CLL. Biological Pointers to the Improved Outcome of Women.

Gender is not widely regarded as a prognostic indicator in CLL. However, the combined data from three MRC randomised trials, CLL1, 2 and 3, and two observational studies for patients with Binet stage A, CLL2A and 3A, over a period of 20 years (1978-1998) totalling 2370 patients, showed a significant survival advantage for women (2p<0.0001). Cox regression analysis of the three randomised trials showed that gender, age, stage and response to therapy were independent prognostic variables. The response to treatment for women was also better than for men receiving the same therapies. The LRF CLL4 randomised trial, which started in 1999 shows the same trend. Preliminary results in 444 patients Binet stages A progressive, B and C showed the same CR/Nod.PR for both sexes (43.5%) but a higher PR rate in women (45%) vs men (36.5%) and a lower proportion of women non-responders to first line therapy (11.5% vs 20%). A number of laboratory investigations in CLL4, which included FISH analysis with five probes, VH mutational status, CD38 and ZAP-70 expression by flow cytometry, showed differences between the sexes, which were significant for 17p and 11q deletions combined and CD38, always in favour of women, as shown in Tables 1 and 2.The clinical and laboratory results suggest that CLL is biologically more benign in women. Women have a lower incidence of CLL, an overall higher incidence of stage A (41.7%) than men (27.3%) in CLL 1, 2, 2A, 3 and 3A and respond better to treatment in all the trials. These differences may be underlined by a higher proportion of 13q del as sole abnormality, a lower proportion of 17p (p53 locus) and 11q deletions and lower levels of CD38. Data on VH mutations and ZAP-70 point in the same direction but the number of cases studied is still small. An additional factor that may play a role in the better outcome for women relates to the effect of oestrogen derivatives which are known to target selectively superoxide dismutase and induce cell kill (Huang et al, Nature 407, 390, 2000).Table 1: FISH analysis by gender (Dohner hierarchical model)Abnormality17p del11q delTrisomy 1213q delOthersMen (286)12%19%10%34%25%Women (94)7%13%11%44%25%p value**0.052* = Combined p value < 0.05 (Chi-Square test)Table 2: Other biological markersCD38 negative (<30%)VH mutatedZAP-70 negativeMen173/335 (52% )51/149 (34%)94/192 (49%)Women68/100 (68%)23/52 (44%)42/68 (62 %)p value<0.005NS0.07

Response to Therapy and Survival in CLL Is Influenced by Genetic Markers. Preliminary Analysis from the LRF CLL4 Trial.

The randomised CLL4 trial was launched in 1999 to answer scientific and management issues. Since then, 730 patients with Binet stages A progressive, B and C have been randomised between 1) chlorambucil (10mg/m2 x 7 days), or 2) fludarabine (FDR) alone (40mg/m2 x 5 d oral or 25mg/m2 x 5 days IV), or FDR plus cyclophosphamide (FDR 24mg/m2 x 5 d oral or 25mg/m2 x 3 d IV plus cyclo 150mg/m2 x 5 d oral or 250mg/m2 x 3 d IV). Information on FISH analysis with 5 probes is available on 440 patients. The incidence of the genetic abnormalities is: 13q del (60%), trisomy 12 (15%), 6q del (8%), 11q del (19%), 17p del (11%); some markers are positive in more than one patient. Analysis by stage showed no differences in the distribution of these markers and analysis by age showed that patients with 13q del were younger (p=0.006) and those with 11q del were older (p=0.01). Analysis by response to all therapies (as the code of individual therapies has not been broken) in 343 patients is shown in Table 1.The results show clearly that almost half the patients with 17p del (p53 locus) are primary non-responders (NR) or show progressive disease (PD) following first line therapy, and have poor survival. 2yr survival of patients with none of these abnormalities was 87% (79%–95%), and using a hierarchical model with normals as the fifth group there was a clear trend in survival between groups (2p<0.00005).When the % of 17p del lymphocytes was correlated with response, it was apparent that the higher the % of 17p del cells the greater the proportion of non-responders (Table 2), with 20% 17p del being the critical threshold.When we grouped together cases with 6q, 11q and 17p deletion plus trisomy 12 (representing 42% of cases) there was a significantly lower response rate compared with the cases without those four abnormalities: CR/Nod PR 35% vs 47%, PR 33% vs 39%, and 32% vs 13%, (Chi-square 13.3; p=0.0003). This study demonstrates that the genetic abnormalities which can be shown in c.80% of patients with active CLL have a significant bearing on the response to treatment and early survival, confirming the survival differences reported by Dohner at al (N Engl J Med 343: 1910–16; 2000). It remains to be determined whether better therapies could overcome the low response rates of patients with poor prognostic genetic abnormalities.Table 1: FISH analysis by response (non-hierarchical) and survivalAbnormalityCR/Nod PRPRNR/PDp value*2 yr survival (95% CI)17p del33%18%48%0.00467% (50%–85%)11q del29%44%27%NS73% (59%–86%)Trisomy 1241%31%29%NS72% (57%–87%)6q del**39%29%32%NS75% (56%–95%)13q del42%37%21%NS91% (86%–95%)* Analysis of individual abnormalities versus the rest; Chi-square test; **326 patientsTable 2: Correlation of response with proportion of 17p del lymphocytes% deletionNo. casesCR/Nod PRPRNR/PD<5 (negative)31043%39%18%5–201861%22%17%>20–100150%13%87%Chi-square 318.8; p<0.0001