Characterizing and prognosticating chronic lymphocytic leukemia in the elderly: prospective evaluation on 455 patients treated in the United States

Chronic lymphocytic leukemia (CLL) accounts for 15 000 diagnosed cases in the USA annually [1]. While incremental improvements in treating CLL have been observed in the past decade [2], the majority of clinical trials leading to these treatment approaches have largely enrolled younger, fitter patients who do not accurately reflect the demographics of CLL patients seen in the community [3–6]. One exception was the CLL-11 study that compared chlorambucil alone to chlorambucil combined with rituximab or obinutuzumab in patients with co-morbidities defined as either a glomerular-filtration rate < 70 mL/min or a cumulative-illness-rating scale ≥ 6 [7]. Other studies have allowed enrollment of elderly patients and performed unplanned subset analyses in an attempt to refine treatments and outcomes in the elderly, but data were inconclusive [8–10]. Moreover, a population-based analysis of 28 590 US patients diagnosed with CLL (1992–2009) showed that the improvement in overall survival (OS) noted in younger patients was less pronounced in the elderly [11]. Furthermore, Brenner et al. [12] showed that improved survival for CLL has not been observed in older patients.

Whether these differences are related to disparities in therapeutic choice, access to care, non-CLL-related deaths in elderly patients, or variations in CLL biology and prognostic indicators is unknown. As the median age of CLL patients at diagnosis approaches 72 years, understanding the biology and outcomes for elderly patients is critical and underscored by the reported inferior survival of these patients.

To examine treatment patterns and disease-related outcomes in elderly CLL patients (defined as ≥ 75 years), we used the Connect® CLL database that enrolled 1494 CLL patients requiring therapy between 2010 and 2014 [13]. These patients were almost entirely enrolled prior to the introduction of novel B-cell receptor (BCR)-targeted therapies. We aimed to establish a benchmark for outcomes in elderly CLL patients treated before the availability of BCR-targeted therapies to help in properly positioning newer agents in the elderly CLL treatment paradigm. Our objective was to compare patient and disease characteristics, prognostic indicators, complications, and disease-related mortality. Further, we aimed to develop a prognostic score that predicts elderly CLL patients at highest risk of CLL- or infection-related deaths. To our knowledge, this represents the largest comprehensive, prospective evaluation of this patient population published to date.

While inferior OS is expected in elderly CLL patients, our analysis of elderly CLL patients treated in a ‘real-world’ setting showed that these patients are more likely to experience CLL- or infection-related deaths. To our knowledge, this has not been reported previously. We developed a prognostic score specifically for this vulnerable patient population, which classified the elderly CLL cohort into high- and low-risk groups with statistical variation in CLL-related mortality.

As the US population ages, identifying optimal therapeutic strategies for the elderly is a critical unmet medical need as few prospective trials have targeted this patient population. Moreover, elderly patients enrolled in clinical trials might not represent the general elderly population treated in the community. While geriatric assessments should be used to provide an objective and comparable measure of elderly status [20], most studies define elderly patients based solely on an age cut-off. As the median age at diagnosis is 72 years, we selected ≥ 75 years of age as the cut-off for this analysis. While there are limitations to selecting an age cut-off, we postulated that a cut-off above the median age at diagnosis would be clinically meaningful. In addition, published prospective data on outcomes for patients who are ≥ 75 years of age are limited [21, 22].

Elderly CLL patients were more often treated with rituximab monotherapy than their younger counterparts who were more likely to receive chemoimmunotherapy [22]. However, the fact that 20% of elderly CLL patients did not receive an anti-CD20 monoclonal antibody is striking, given that all patients were treated after 2010. Even in the younger cohort, we observed that 10% of patients did not receive any anti-CD20 antibodies. To better understand this variation, we assessed whether patterns of care differed based on health insurance coverage or geographic location of the treating institution. Elderly CLL patients were less likely to receive rituximab-based therapies than younger patients, regardless of insurance provider. However, patients residing in the South were more likely to receive anti-CD20 therapy compared with patients living on the West coast. A comparable observation was reported in a study of follicular lymphoma patients in the West of the USA who were less likely to receive rituximab-based maintenance therapy [23]. This may reflect differences in the treating institution and/or setting. Rituximab use has increased in hospitals while declining in clinics, which could account for the imbalance in treatment between geographic locations [24]. These data suggest that real-world findings differ from clinical trial observations.

Regardless of LOT, responses appeared lower in elderly CLL patients. Although responses were assessed by treating physicians and were not centrally reviewed, CR in the younger patients at LOT1 (42.3%) was comparable to the response (44%) reported for treatment-naïve patients in the CLL-8 trial of rituximab plus fludarabine/cyclophosphamide [6]. Only 25.9% of elderly CLL patients achieved a CR in LOT1. Given the association between survival and the depth of remission [25], this finding is critical and might contribute to the inferior outcomes noted in our elderly cohort.

Despite the typically indolent nature of CLL, we observed critical outcome differences at a median follow-up of 32.6 months. OS was inferior in elderly CLL patients in any LOT group. Given the predictably inferior OS in the elderly due to competing co-morbidities and deaths from other causes, we compared CLL-related deaths between both groups in LOT1 and LOT ≥ 2. Only 5% of patients < 75 years in LOT1 experienced CLL-specific deaths while 13% of elderly CLL patients died from CLL alone. This difference was statistically significant (p = 0.0005). A similar observation was noted in LOT ≥ 2 (31% for ≥ 75 years vs. 22% for < 75 years; p = 0.0277). Since infections are a major cause of CLL-related deaths, we evaluated the differences in deaths due to CLL or infection in both LOT groups. The difference remained significant (p < 0.0001 in LOT1; p = 0.0014 in LOT ≥ 2).

We subsequently studied prognostic indicators for CLL- or infection-related deaths in elderly CLL patients. We identified three factors that were significant in a multivariable analysis: time from diagnosis to therapy initiation of < 3 months, enrollment therapy other than BR, and anemia. While a time from diagnosis to therapy of < 3 months may suggest patients had more aggressive disease, this is not necessarily related to disease staging. Indeed, the majority of patients in each LOT and age group had Rai stage 0–2. Rai stage did also not differ significantly between younger and older patients. The prognostic score was used to classify elderly CLL patients according to high- or low-risk of CLL-related death (30.6 vs. 10.3%, respectively; p = 0.0002). Contrary to the prognostic models published by Pflug et al. [26], and The International Prognostic Index for patients with CLL (CLL-IPI) working group [27] in which all patients were included regardless of age, our score was specifically designed for elderly CLL patients. Notably, Pflug et al. [26] and the CLL-IPI working group [27] identified older age as an independent factor negatively impacting survival. Our model is also specific to patients receiving therapy as patients under observation were not enrolled to the registry.

Several limitations inherent in any registry-based observational study were encountered during our study. These include the non-random allocation of patients to specific interventions, the assessment of outcomes by non-blinded individuals, and the greater potential for missing data [28]. In the Connect CLL registry, responses were not centrally assessed and indications to treat were based on the treating physician’s judgment. Comprehensive molecular and cytogenetic evaluation was missing for some patients. Our analysis also has limitations that are specific to the Connect CLL registry. Only patients requiring therapy were enrolled in the registry. Patients who died without starting therapy were excluded. The registry predates the introduction of BCR-targeted therapies; therefore, the patients in this registry were not treated with these novel agents. As with any registry, patients were treated with a number of different therapies. The small size of the cohort and the inclusion of only 181 patients in the prognostic model may also be limiting factors. However, despite the small sample size we believe that these results are meaningful as they relate specifically to elderly patients. Importantly, these data also represent the largest US population of CLL patients treated outside of interventional clinical trials in the chemoimmunotherapy era.

Our finding of increased mortality related to elderly CLL patients highlights the urgent need for therapies tailored to this population and underscores the need to refine CLL treatment for the elderly as current therapies and strategies appear suboptimal. This might reflect a limited enrollment of elderly patients into clinical trials and highlight a flaw in the assumption that effective regimens in younger patients will be effective in elderly patients. As new BCR-targeted agents are increasingly used, their role in elderly CLL patient treatment will require critical analysis to balance efficacy with toxicity. Our data on CLL- and infection-related mortality using traditional therapies are a benchmark against which novel therapies can be measured. Finally, the proposed prognostic score, while requiring validation in patients treated with BCR-targeted therapies, could be used to stratify elderly CLL patients on their enrollment into future clinical trials.

These data represent the real-world experiences of a large population of CLL patients treated across the USA. Within the limitations of an observational registry we have shown that elderly CLL patients have inferior outcomes with a cumulative increased risk of death from CLL regardless of LOT. Recent improvements in survival for younger patients with CLL have still to be achieved in elderly CLL patients. While elderly people have increased mortality versus younger people regardless of CLL status, it will be important to identify new therapies that can improve the outcomes for elderly CLL patients, similar to the advances seen in younger CLL patients. This unique prognostic model for patients ≥ 75 years could identify those patients who would benefit from early treatment or treatment with novel therapies.