Abstract
A ‘watch-and-wait’ strategy is recommended for most patients with early-stage chronic lymphocytic leukemia (CLL) prior to treatment initiation. In the Connect® CLL registry, a prospective observational cohort study of 1494 patients treated in 199 US centers, median time to first-line treatment initiation was 3.8, 1.5, and 0.6 years for patients with Rai stage 0, 1, and ≥2, respectively. Only 60% of patients with Rai stage 0/1 underwent FISH/cytogenetic testing prior to initiation of a new line of therapy. Lymphocytosis and lymphadenopathy were the most common reasons for treatment initiation. Lymphocytosis as a reason for treatment initiation was associated with inferior event-free survival at Rai stage 0/1. Short treatment duration was associated with inferior overall survival regardless of Rai stage; sensitivity analyses confirmed the association. The Connect CLL registry provides valuable information on a real-world population of patients with CLL, clarifying both the timing and rationale for initiating therapy.
Introduction
Chronic lymphocytic leukemia (CLL) is the most common leukemia in adults in the USA, with an estimated 18,960 new cases and 4660 deaths in 2016 [Citation1]. Median survival from diagnosis varies from 18 months to more than 10 years, with prognosis depending on the presence of several clinical and cytogenetic features [Citation2]. The Rai [Citation3] and Binet [Citation4] staging systems have historically been used to assess prognosis and guide management in patients with CLL. The majority of patients with newly diagnosed CLL have lower-risk disease (Rai stage 0/1), experience an indolent clinical course, and can be observed from diagnosis until disease progression necessitates treatment initiation [Citation2,Citation5,Citation6]. The potential survival benefit of early intervention remains to be proven [Citation6].
Guidelines list multiple criteria to define active disease and the need for therapy [Citation5,Citation6], and to suggest treatment options based on factors including clinical stage, patient fitness and comorbidities, and risk factors [Citation2,Citation6]. However, adoption of clinical guidelines into routine practice is often suboptimal [Citation7,Citation8]. Furthermore, the approval of several new oral drugs, and additional molecular prognostic data and models, such as the CLL International Prognostic Index (IPI) [Citation9], can make individual treatment decisions more challenging [Citation10]. Reasons for initiating therapy and the choice of first-line therapy offered to patients with CLL in community practices in the USA have not been fully investigated. Understanding these treatment patterns in patients with Rai stage 0/1 CLL is essential to reduce unnecessary treatment and to improve outcomes.
Real-world data from observational cohort studies can improve our understanding of practice patterns outside of clinical trials and the data collected may be more representative of the general CLL population. The Connect® CLL Registry was established with a primary objective of observing treatment patterns, and understanding the rationale behind these treatment patterns, for patients with CLL treated in the community across the USA. The objectives of this current study were to describe the reasons for initiating first-line therapy in patients with Rai stage 0/1 CLL, and to evaluate predictors of survival in these patients.
Methods
Patients and methods
The Connect CLL Registry (NCT01081015), a large, US-based, multicenter, prospective observational cohort study that enrolled patients with CLL at the initiation of first or subsequent therapy, has been described previously [Citation11]. Study centers with experience in oncology/hematology trials and registries, as well as an adequate number of patients with CLL, were selected to participate. The Connect CLL Registry is non-interventional, with all scheduled visits, treatment selections, and disease response assessments performed at the discretion of the treating physician in accordance with their usual care. Participation in the registry was voluntary, and patients could withdraw at any time without affecting their subsequent medical care. Sites were encouraged to enroll all eligible patients as they presented to their physician; each center could enroll up to 30 patients.
Eligible patients were ≥18 years of age, with CLL as defined by the International Workshop on CLL (IWCLL) guidelines [Citation5], and had initiated therapy within 2 months prior to study enrollment. Fluorescence in situ hybridization (FISH)/cytogenetic test results for del(17p) testing were centrally reviewed. The study protocol was approved by a Central Institutional Review Board (IRB) (Quorum Review IRB, Seattle, WA, USA), and was used by 100 centers for site-specific approval; the remaining centers gained approval from their local IRB. All patients provided written informed consent.
Statistical analyses
Patients enrolled in the Connect CLL Registry initiating their first line of therapy were stratified according to Rai stage at diagnosis (Rai stage 0, 1, or ≥2). Rai stage and reasons for initiating therapy were reported by the treating physician. Disease response was recorded as complete response (CR), partial response (PR), stable disease, or progressive disease as determined by the treating physician. Response assessment was reported by the investigator utilizing whatever methods they deemed appropriate. Timing and methods of establishing treatment response may have differed across sites; response assessments will only be collected when performed as part of normal practice. Descriptive statistical methods were used to assess demographics and baseline characteristics, reasons for initiating treatment, treatments selected, and treatment outcomes. The data cutoff date for the analyses was 22 February 2017.
Univariate and multivariable logistic regression analyses were used to evaluate predictors of first-line overall response rate (ORR) and CR. More than 20 potential predictors were analyzed, including age, race, gender, treatment center, Charlson Comorbidity Index (CCI) score [Citation12], prior malignancies, enrollment therapy, del(17p) and CD38 abnormalities, time from diagnosis to treatment initiation, and reason for treatment initiation. Progressive lymphocytosis was defined as lymphocytosis with an increase of 50% lymphocytes or a lymphocyte doubling time of <6 months. Progressive lymphadenopathy was defined as an increase in the number of regions with enlarged lymph nodes (i.e. >5 cm in the longest diameter) or symptomatic lymphadenopathy as assessed by the investigator. Variables significant at the p < .15 level, as identified in the univariate analysis, were included in the multivariable logistic regression analysis. In the multivariable analysis, stepwise and chi-square score selection methods were used to identify independent significant predictors. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.
Event-free survival (EFS; ‘event’ defined as the first occurrence of death, progression/relapse, or transformation) and overall survival (OS) were measured from study enrollment. Patients who did not experience an event were censored at the data cutoff date of 22 February 2017. The Kaplan–Meier product-limit method [Citation13] was used to estimate EFS and OS by Rai stage; p values from the log-rank test for comparison of survival distributions were provided. Univariate and multivariable Cox regression analyses were used to evaluate predictors of EFS and OS. Variables identified in the univariate Cox regression analysis were used to identify independent significant predictors in the multivariable Cox regression analysis; the results were verified using subgroup scores to create the parsimonious model. Hazard ratios (HRs) and 95% CIs were calculated.
Missing data were not imputed as per registry protocol. Missing data were assumed to be missing at random, and sensitivity analyses were conducted where appropriate. All statistical significance was assessed at a 5% level (two-sided). All statistical analyses were performed using SAS® version 9.2 statistical software (SAS Institute, Cary, NC).
Results
From March 2010 to January 2014, 1494 patients were enrolled into the Registry from 179 community (n = 1311), 17 academic (n = 155), and 3 government (n = 28) centers across the USA. Overall, 889 (60%) patients were enrolled within 60 days of initiation of the first line of therapy. Of these, 684 patients had available Rai stage classification; 172 (25%) had Rai stage 0 CLL; 191 (28%) had Rai stage 1, and 321 (47%) had Rai stage ≥2 ().
Figure 1. Line of therapy and Rai stage at enrollment. CLL: chronic lymphocytic leukemia; LOT: line of therapy.
Demographics and baseline characteristics are shown in . Patient characteristics were similar across the Rai groups and the overall study population, except median time from diagnosis to treatment initiation, which was longer in patients with Rai stage 0 compared with Rai stage 1 and Rai stage ≥2. Based on historical information collected at baseline, the median time to treatment initiation was 3.8 years (range 0–20) and 1.5 years (range 0–16) for patients with Rai stage 0 and 1, respectively. At enrollment, FISH and/or cytogenetic testing was recorded in 105 (61%) patients with Rai stage 0; of these, 73 (70%) had abnormal FISH or cytogenetic test results. FISH/cytogenetic testing was performed in 115 (60%) patients with Rai stage 1; of these 86 (75%) had abnormal FISH/cytogenetic test results. Frequency of del(11q), del(13q), del(17p), and trisomy 12 are shown in .
Table 1. Baseline characteristics at enrollment for patients receiving first line of therapy.
Reasons for treatment initiation and initial therapy
The most common reasons for initiating first-line treatment in patients with Rai stage 0/1 at diagnosis were: massive bulky or progressive lymphadenopathy (40%), progressive lymphocytosis (39%), and bone marrow failure (28%); however, more than one reason for initiating treatment could be given and 42% of patients had more than one reason for initiating therapy (). Lymphocytosis was given as the only reason for treatment initiation in 18% of patients. The most commonly used first-line therapy for patients with Rai stage 0 and Rai stage 1 in this population was fludarabine + cyclophosphamide + rituximab (FCR; 28 and 30%, respectively) (). Other common therapies were bendamustine + rituximab (BR; 19 and 21%) and rituximab monotherapy (14 and 12%, respectively).
Table 2. Reasons for initiating therapy for patients with Rai stage 0/1 receiving first line of therapy.
Table 3. Type of enrollment therapy by Rai staging (most frequent first-line regimens only).
Response
Of the 363 patients with Rai stage 0/1, a total of 344 (95%) were assessed for response; 19 (5%) had no response assessment recorded. Overall, 257 patients experienced either a PR or a CR (ORR = 75%), and 166 patients (48%) experienced a CR.
Two factors were independent significant predictors of CR in the multivariable analysis (p < .05): enrollment therapy with BR or FCR, yes versus no (OR 3.33, 95% CI 2.11–5.24), and thrombocytopenia as a reason for treatment initiation, no versus yes (OR 1.97, 95% CI 1.01–3.81). There were three independent significant predictors of ORR (p < .05): enrollment therapy with BR or FCR, yes versus no (OR 3.57, 95% CI 1.70–7.52); del(17p), no versus yes (OR 4.57, 95% CI 1.56–13.41); and treatment duration >3 months versus ≤3 months (OR 3.11, 95% CI 1.53–6.33).
As treatment duration was limited for patients who discontinued early, a sensitivity analysis was performed that only included 316 (92%) patients who remained on study for ≥3 months. This analysis confirmed the findings of the main analysis, with all three predictors of ORR maintaining statistical significance. As part of the sensitivity analyses, patients treated for >3 months were compared with those treated for ≤3 months. No significant differences were observed between the groups with respect to gender, race, and CCI score; however, patients treated for ≤3 months were older (median age 71 versus 66 years, p < .001).
Event-free survival
After a median follow-up of 48.3 months for all 684 patients initiating the first line of therapy, 185 (51%) of the 363 patients with Rai stage 0/1 experienced an EFS event. Event rate was similar between patients with Rai stage 0 and Rai stage 1 (50 versus 52%, p = .723). Kaplan–Meier plots of EFS are shown in . Median EFS was similar in patients across Rai stage 0, 1, and ≥2. Multivariable Cox regression analysis identified three independent predictors of inferior EFS in patients with Rai stage 0/1 (p < .05): age ≥75 years (HR 1.91; p < .0001); enrollment therapy other than BR or FCR (HR 1.49; p = .009); and lymphocytosis as a reason for treatment initiation (HR 1.63; p = .001). Del(17p) was not found to be a significant predictor of EFS in univariate analyses (HR 0.69; p = .24).
Figure 2. Kaplan–Meier EFS curves for patients enrolled at the first line of therapy, stratified by Rai stage. CI: confidence interval; EFS: event-free survival.
Similar analysis of patients with Rai stage ≥2 identified the following independent predictors of inferior EFS: del(17p), yes versus no (HR 2.51; p < .003); del(13q), no versus yes (HR 1.99; p < .004); enrollment therapy other than BR or FCR (HR 1.78; p = .014); insurance other than private (HR 1.72; p = .025); lymphadenopathy as a reason for treatment initiation (HR 1.80; p = .012); and treatment duration ≤3 months versus >3 months (HR 1.67, p = .029). Sensitivity analyses performed in patients who remained on study for >3 months confirmed the above findings.
Overall survival
Kaplan–Meier plots of OS are shown in . After a median follow-up of 48.3 months, 38 (22%) of 172 patients with Rai stage 0 and 29 (15%) of 191 patients with Rai stage 1 enrolled at first line of therapy had died. Cause of death included CLL progression (excluding Richter transformation) (24 versus 41% in Rai stage 0 and 1, respectively), second primary malignancy (SPM) (21 versus 14%), infection (11 versus 7%), and cardiac events (11 versus 7%). Of 321 patients with Rai stage ≥2 at first line of therapy, 73 (23%) died. Cause of death included CLL progression (excluding Richter transformation) (26%), infection (18%), SPM (8%), and cardiac events (8%).
Figure 3. Kaplan–Meier OS curves for patients enrolled at the first line of therapy, stratified by Rai stage. CI: confidence interval; OS: overall survival.
Multivariable Cox regression analysis found three independent significant predictors of inferior OS in patients with Rai stage 0/1: age ≥75 years (HR 3.89; p < .0001); CCI score ≥3 versus 2 (HR 1.86; p = .007); and therapy duration ≤3 months versus >3 months (HR 1.71; p = .020). Sensitivity analyses confirmed these results.
Multivariable analysis of patients with Rai stage ≥2 identified the following independent predictors of inferior OS: insurance other than private (HR 2.07; p = .004); treatment duration ≤3 months versus >3 months (HR 2.08, p = .001); CD38 abnormality, yes versus no (HR 1.60, p = .041); and race other than white (HR 2.14, p = .008). Sensitivity analyses in patients who stayed on study for >3 months confirmed the above results; however, race was no longer significant (p = .157). Del(17p) was not found to be a significant predictor of OS in univariate analyses (HR 0.65; p = .28).
Discussion
To our knowledge, the Connect CLL registry is the single largest prospective study of patients with CLL examining academic and community practice patterns across the USA. In this analysis, we found that patients with Rai stage 0 initiated first-line therapy 3.8 years after diagnosis while patients with Rai stage 1 initiated first-line therapy 1.5 years after diagnosis.
We also believe that this is the largest study to investigate the reasons for initiating therapy in patients initially diagnosed with Rai stage 0/1 CLL. We show that the most common reasons for starting treatment in these patients were lymphadenopathy and lymphocytosis. In general, newly diagnosed patients with Rai stage 0, and some patients with Rai stage 1, can be monitored without therapy. However, therapy should be initiated when there is evidence for progressive or symptomatic disease [Citation5]. Patients with high-count monoclonal B-cell lymphocytosis (MBL) were not included in this registry. Differentiating between patients with high-count MBL and Rai stage 0 remains difficult, as many of the biological characteristics are similar [Citation14]. The threshold for defining high-count MBL (5 × 109 B cells/L) has led to the reclassification of many patients with Rai stage 0 as having high-count MBL [Citation15]. In this study, progressive lymphocytosis was often provided as a reason for treatment initiation; however, current guidelines do not recommend treatment initiation for lymphocytosis alone [Citation16]. While it is possible that some patients with MBL were misclassified as Rai stage 0 CLL and some patients received treatment for lymphocytosis alone before it was warranted, many patients had more than one reason for initiating therapy, with a median time from diagnosis to treatment initiation of more than 3 years. These data likely provide the clearest representation of common practices for initiating therapy for Rai stage 0/1 CLL in the USA.
The most commonly used first-line therapies for patients with Rai stage 0/1 were FCR and BR, as recommended by guidelines for patients eligible for intensive chemoimmunotherapy [Citation2,Citation6,Citation16]. Surprisingly, a number of patients did not receive rituximab-containing therapy, despite being treated since the introduction of rituximab as first-line therapy. Studies have shown that FCR improves treatment outcomes, including response rates, progression-free survival (PFS) and OS, compared with fludarabine + cyclophosphamide [Citation17–21]. The safety and efficacy of BR in patients with previously untreated CLL have also been demonstrated [Citation22]. The phase 3 CLL10 trial comparing FCR versus BR, found higher CR rates and longer PFS in the FCR arm. Based on these data, FCR is currently considered the standard first-line chemoimmunotherapy regimen for patients under the age of 65 years who are physically fit. However, in elderly patients, there was a higher rate of toxicity and infection, and no improvement in PFS with FCR versus BR [Citation23]. Our registry enrolled patients prior to the introduction of B-cell receptor (BCR)-targeting therapies, such as ibrutinib and idelalisib. As ibrutinib has since been approved as first-line therapy for patients with del(17p), it will be interesting to assess how treatment patterns and outcomes for patients initially diagnosed with Rai stage 0/1 change in the era of BCR-targeted therapies.
Multivariable regression analysis showed that older age (≥75 years) was an independent predictor of both inferior EFS and OS for patients with Rai stage 0/1. Interestingly, older age was not predictive of survival at Rai stage 2. Enrollment therapy other than BR or FCR was predictive of inferior EFS, regardless of Rai stage, highlighting the importance of first-line treatment selection. SPM was reported as the cause of death in 18% of patients. This is higher than the incidence of SPMs reported in French CLL registries (11.5%) [Citation24]. Although it was not possible to determine how many of these SPMs were therapy-related, an association between CLL treatment and SPMs has been reported previously [Citation25,Citation26]. While del(17p) was a significant predictor of ORR in all patients and of EFS in patients with Rai stage ≥2, it was not found to be a significant predictor of EFS in patients with Rai stage 0/1 or of OS. This may be due to incomplete or missing del(17p) data for some patients. Multivariable analysis of prognostic factors was further limited by the lack of data on IGHV and TP53 mutation status.
Duration of first-line therapy was found to be a significant predictor of EFS and OS, with longer duration (≥3 months) associated with improved survival. Patients who had treatment duration <3 months were also less likely to respond to treatment. This may be caused by selection bias as patients with refractory disease or those who experience severe treatment-related adverse events will have a shorter duration of therapy. Those patients who stay on therapy will also inherently have longer OS. However, sensitivity analyses showed that while patients with treatment duration <3 months were older than those patients with treatment duration of ≥3 months, no other significant differences in patient characteristics were observed. These data provide insights into treatment selection and patterns of care associated with improved outcomes.
As expected, the interval between diagnosis and therapy initiation was much longer for patients with Rai stage 0 (3.8 years) than for those with Rai stage 1 (1.5 years) and Rai stage ≥2 (0.6 years). This is in line with the current ‘watch-and-wait’ strategy. However, in patients with early-stage CLL, this strategy may be challenged by the emergence of highly active targeted drugs such as ibrutinib and idelalisib [Citation27,Citation28]. The interval of 3.8 years from diagnosis to treatment for patients with Rai stage 0 is shorter than that reported in other studies [Citation29,Citation30]. In a study of patients treated at a single academic center, median time to first therapy was 6.1 years for patients treated by a hematologist, increasing to 9.2 years for patients treated by a CLL specialist hematologist [Citation29]. This indicates that many patients treated predominantly in community-based practice settings are treated earlier than might be expected under the care of a CLL specialist. These data suggest a role for evaluation and treatment planning by a CLL specialist even for Rai stage 0/1 patients who do not require immediate treatment.
Despite recommendations [Citation2,Citation5,Citation16], FISH and/or cytogenetic testing was only performed in approximately 60% of patients with Rai stage 0/1 before treatment, indicating a potential unmet need for patients with CLL in the community and an educational opportunity for physicians. Most patients in this study (73%) had abnormal FISH or cytogenetic test results, including del(11q) (18%) and del(17p) (10%). While these high risk cytogenetic characteristics are not a specific indicator for therapy, they are generally considered to occur later in the disease course and to predict a worse outcome [Citation10,Citation31,Citation32]. Given that novel treatment strategies now exist based on FISH results, FISH testing should be performed in all patients prior to the start of treatment.
Information on the reasons for, and timing of, treatment initiation may be relevant in the development of novel prognostic scoring systems and identification of asymptomatic patients who may benefit from early, rather than delayed, treatment. New prognostic scoring systems using combinations of clinical and/or genetic markers are currently under development; these may be particularly valuable in lower-risk patients, where the current Rai and Binet staging systems are less useful [Citation33–35].
The main limitations of the current study are those generally associated with any observational cohort study, e.g. the risk of selection bias, and assessment of treatment effectiveness in the absence of controls or randomization. Additionally, responses were not centrally assessed, and no formal response criteria were mandated per protocol. Missing data or underreporting of critical data like performance status, diagnostic testing, and disease response may also be a source of bias. In order to minimize this bias, participating centers were reminded to invite every eligible patient to participate in the registry. Data were also reviewed on a regular basis to ensure accurate reporting and sensitivity analyses conducted to compensate for these biases. A number of caveats specific to this analysis should also be noted including the heterogeneous nature of the treatments prescribed and the lack of novel regimens such as ibrutinib. Information on prognostic factors such as FISH data and IGHV mutation status was also missing in many patients. Despite these limitations, the Connect CLL Registry is one of the largest prospective cohort studies of CLL patients including patients from multiple geographically diverse and predominantly community-based centers. Data from disease registries also provide valuable information that is more representative of the general population than patients enrolled in clinical trials. Such real-world data offer a highly relevant perspective on patient experiences and disease outcomes. As such, taking data from registries and clinical trials together can provide a complementary view of CLL patients undergoing therapy in the USA.
Potential conflict of interest
Disclosure forms provided by the authors are available with the full text of this article online at https://doi.org/10.1080/10428194.2018.1427860.
ICMJE Forms for Disclosure of Potential Conflicts of Interest
Download Zip (16.4 MB)Acknowledgments
The authors received medical writing support in the preparation of this article from Eva Polk, PhD, CMPP, and Victoria Edwards, PhD, of Excerpta Medica BV, funded by Celgene Corporation. The Connect® CLL Scientific Steering Committee acknowledges the contributions of all past and current members of the committee for their guidance in the design of the registry and participation in the analysis of the data, including Matthew Davids, Charles Farber, Ian Flinn, Christopher R. Flowers, David L. Grinblatt, Neil E. Kay, Michael Keating, Thomas J. Kipps, Mark F. Kozloff, Nicole Lamanna, Susan Lerner, Anthony Mato, Chadi Nabhan, Chris L. Pashos, Jeff P. Sharman, and Mark Weiss.
Additional information
Funding
References
- American Cancer Society (ACS) [Internet]. Georgia: ACS. Chronic lymphocytic leukemia. [cited 2016 May]. Available from: http://www.cancer.org/acs/groups/cid/documents/webcontent/003111-pdf.pdf
- Eichhorst B, Dreyling M, Robak T, et al. Chronic lymphocytic leukemia: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2011;22(Suppl 6):vi50–vi54.
- Rai KR, Sawitsky A, Cronkite EP, et al. Clinical staging of chronic lymphocytic leukemia. Blood. 1975;46:219–234.
- Binet JL, Auquier A, Dighiero G, et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer. 1981;48:198–206.
- Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood. 2008;11:5446–5456.
- Hallek M. Chronic lymphocytic leukemia: 2015 Update on diagnosis, risk stratification, and treatment. Am J Hematol. 2015;90:446–460.
- Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999;282:1458–1465.
- Sahota IS, Kostaras X, Hagen NA. Improving access to cancer guidelines: feedback from health care professionals. Curr Oncol. 2015;22:392–398.
- The International CLL-IPI Working Group. An International Prognostic Index for patients with chronic lymphocytic leukaemia (CLL-IPI): a meta-analysis of individual patient data. Lancet Oncol. 2016;17:779–790.
- Hallek M. Signaling the end of chronic lymphocytic leukemia: new frontline treatment strategies. Blood. 2013;122:3723–3734.
- Mato A, Nabhan C, Kay NE, et al. Real-world clinical experience in the Connect® chronic lymphocytic leukaemia registry: a prospective cohort study of 1494 patients across 199 US centres. Br J Haematol. 2016;175:892–903.
- Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383.
- Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457–481.
- Morabito F, Mosca L, Cutrona G, et al. Clinical monoclonal B cell lymphocytosis versus Rai 0 chronic lymphocytic leukemia: a comparison of cellular, cytogenetic, molecular, and clinical features. Clin Cancer Res. 2013;19:5890–5900.
- Strati P, Shanafelt TD. Monoclonal B-cell lymphocytosis and early-stage chronic lymphocytic leukemia: diagnosis, natural history, and risk stratification. Blood. 2015;126:454–462.
- National Comprehensive Cancer Network [Internet]. Fort Washington, PA: NCCN. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines 2017). Non-Hodgkin’s lymphomas. Version 1. 2017. [cited 2017 Aug 9]. Available from: https://www.nccn.org/professionals/physician_gls/f_guidelines.asp
- Fischer K, Bahlo J, Fink AM, et al. Long-term remissions after FCR chemoimmunotherapy in previously untreated patients with CLL: updated results of the CLL8 trial. Blood. 2016;127:208–215.
- Hallek M, Fischer K, Fingerle-Rowson G, et al. Addition of rituximab to fludarabine and cyclophosphamide in patients with chronic lymphocytic leukaemia: a randomised, open-label, phase 3 trial. Lancet. 2010;376:1164–1174.
- Robak T, Dmoszynska A, Solal-Céligny P, et al. Rituximab plus fludarabine and cyclophosphamide prolongs progression-free survival compared with fludarabine and cyclophosphamide alone in previously treated chronic lymphocytic leukemia. JCO. 2010;28:1756–1765.
- Tam CS, O’Brien S, Wierda W, et al. Long-term results of the fludarabine, cyclophosphamide, and rituximab regimen as initial therapy of chronic lymphocytic leukemia. Blood. 2008;112:975–980.
- Keating MJ, O’Brien S, Albitar M, et al. Early results of a chemoimmunotherapy regimen of fludarabine, cyclophosphamide, and rituximab as initial therapy for chronic lymphocytic leukemia. JCO. 2005;23:4079–4088.
- Fischer K, Cramer P, Busch R, et al. Bendamustine in combination with rituximab for previously untreated patients with chronic lymphocytic leukemia: a multicenter phase II trial of the German Chronic Lymphocytic Leukemia Study Group. JCO. 2012;30:3209–3216.
- Eichhorst B, Fink AM, Bahlo J, et al. First-line chemoimmunotherapy with bendamustine and rituximab versus fludarabine, cyclophosphamide, and rituximab in patients with advanced chronic lymphocytic leukaemia (CLL10): an international, open-label, randomised, phase 3, non-inferiority trial. Lancet Oncol. 2016;17:928–942.
- Cornet E, Jégu J, Mounier M, et al. Second cancer incidence among chronic lymphocytic leukemia (CLL) patients: a French population-based study. Blood. 2014;124,abstract 3303.
- Guidez S, Demarquette H, Nudel M, et al. Impact of anti-CD20 antibodies on the incidence of second primary cancers in patients treated for chronic lymphocytic leukemia. Blood. 2015;126:abstract 5291.
- Maurer C, Langerbeins P, Bahlo J, et al. Effect of first-line treatment on second primary malignancies and Richter’s transformation in patients with CLL. Leukemia. 2016;30:2019–2025.
- Burger JA, Tedeschi A, Barr PM, et al. Ibrutinib as initial therapy for patients with chronic lymphocytic leukemia. N Engl J Med. 2015;373:2425–2437.
- Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in relapsed chronic lymphocytic leukemia. N Engl J Med. 2014;370:997–1007.
- Shanafelt TD, Kay NE, Rabe KG, et al. Hematologist/oncologist disease-specific expertise and survival: lessons from chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL). Cancer. 2012;118:1827–1837.
- Shanafelt TD, Rabe KG, Kay NE, et al. Age at diagnosis and the utility of prognostic testing in patients with chronic lymphocytic leukemia. Cancer. 2010;116:4777–4787.
- Döhner H, Stilgenbauer S, James MR, et al. 11q deletions identify a new subset of B-cell chronic lymphocytic leukemia characterized by extensive nodal involvement and inferior prognosis. Blood. 1997;89:2516–2522.
- Döhner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med. 2000;343:1910–1916.
- Morabito F, Cutrona G, Gentile M, et al. Definition of progression risk based on combinations of cellular and molecular markers in patients with Binet stage A chronic lymphocytic leukaemia. Br J Haematol. 2009;146:44–53.
- Gentile M, Cutrona G, Mosca L, et al. Prospective validation of a risk score based on biological markers for predicting progression free survival in Binet stage A chronic lymphocytic leukemia patients: results of the multicenter O-CLL1-GISL study. Am J Hematol. 2014;89:743–750.
- Pflug N, Bahlo J, Shanafelt TD, et al. Development of a comprehensive prognostic index for patients with chronic lymphocytic leukemia. Blood. 2014;124:49–62.