Real-world disease characteristics and frontline treatments used in chronic lymphocytic leukaemia in Bulgaria: An observational cohort study (DESCRIBE)

Abstract

Objective

Bruton tyrosine kinase inhibitors and B-cell lymphoma 2 inhibitors are currently the new standard active agents in chronic lymphocytic leukaemia. This observational study describes the real-world use of frontline treatment in Bulgarian patients with active chronic lymphocytic leukaemia and characterises the patient population receiving these treatments.

Methods

DESCRIBE was a physician-chart review study conducted at eight haematology centres across Bulgaria. Adult patients with confirmed active chronic lymphocytic leukaemia requiring frontline treatment initiation after January 2022 were considered for enrolment.

Results

A total of 90 patients were included in the study. The mean age (standard deviation) was 62.8 (11.2) years, and 58% were male. The median (range) duration of disease was 6 (0–198) months at the time of frontline treatment initiation and 19.2 (1.2–206.4) months at enrolment. Comorbidities were reported in 58 (64%) patients, with cardiovascular diseases being the most common (57%). Cytogenetic testing rates were lower, particularly for del(11q), tumour protein p53 gene and immunoglobulin heavy chain variable region gene mutation status, which were assessed in only 44%, 21% and 37% of patients, respectively. At the time of frontline treatment initiation, most patients presented with Rai stage II (44%) or Binet stage B (63%). Targeted therapy was used in 77% of cases (46% Bruton tyrosine kinase inhibitors and 31% B-cell lymphoma 2 inhibitors), while chemoimmunotherapy was used in the remaining 23%. Patients with high-risk chronic lymphocytic leukaemia were more likely to receive second-generation Bruton tyrosine kinase inhibitors (odds ratio = 5.59, p < 0.001), and each additional high-risk trait further increased the odds of its use (odds ratio = 2.35, p < 0.01).

Conclusions

Our findings indicate a shift towards the adoption of novel therapies in Bulgaria, with Bruton tyrosine kinase inhibitors and B-cell lymphoma 2 inhibitors now dominating the contemporary frontline treatment landscape (2022–2023). Improving the rate of testing for specific high-risk molecular and genetic markers is vital to enhancing outcomes in patients with chronic lymphocytic leukaemia. In the era of proven efficacy of targeted agents, chemoimmunotherapy should be discouraged, in line with international guidelines.

Introduction

With an annual incidence of 4.2/100,000, ¹ chronic lymphocytic leukaemia (CLL) is the most common type of leukaemia in European and North American adults. ² Over the past three decades, the burden of CLL has increased significantly, with an estimated annual incidence growth of 1.86 globally and 2.85 in Eastern Europe.^2,3

For patients with active or symptomatic CLL, primary treatment objectives focus on improving quality of life and extending overall survival. These outcomes are influenced not only by the efficacy of therapy and the sequence in which treatments are administered but also by economic factors and the characteristics of healthcare systems in each region.^1,2,4

Recent advances in the understanding of CLL biology have led to substantial changes in treatment paradigms, with traditional chemoimmunotherapy (CIT) approaches increasingly replaced by targeted therapies (TTs), which have demonstrated marked improvements in survival rates.^5–9 Currently, frontline therapies (FLTs) predominantly use inhibitors targeting Bruton tyrosine kinase (BTK) and B-cell lymphoma 2 (BCL2) proteins, which are now regarded as the standard for active anti- leukaemic treatment. ⁵ According to clinical guidelines, patients with active disease may receive either first- or second-generation BTK inhibitors (BTKi), administered continuously until disease progression, or a time-limited regimen combining a BCL2 inhibitor (BCL2i) with a CD20 monoclonal antibody or a combination of BTKi and BCL2i.^1,10,11 At present, the role of CIT in CLL treatment is extremely limited and is reserved for select patient populations, such as when access to TT is restricted or in cases of rare patient-specific contraindications.^7,10 The decision to initiate CLL treatment is based on the presence of active or symptomatic disease, characterised by progressive bone marrow failure, progressive splenomegaly/adenopathy, rapid lymphocyte doubling, B symptoms, extranodal involvement or autoimmune complications. ¹¹ However, selecting the appropriate treatment strategy is more complex and depends on multiple factors. Conventionally, these factors are divided into two categories: factors related to the patient and disease (e.g. clinical stage, genetic risk factors, age, comorbidities, known side effect profiles, patient preference and adherence) and factors related to currently approved medicines for CLL treatment.^1,4,10,11

In Bulgaria, a Central Eastern European country with a population of approximately 6.5 million, ¹² 639 new leukaemia cases were reported in 2022. ¹³ Limited data are available on CLL epidemiology because the National Cancer Registry has been inactive since 2015, when 163 new CLL cases were recorded ¹⁴ —a number presumed to have increased over time. This increase may be related to improved diagnostic options, better patient access to healthcare and an ageing population. Based on historical registry data and European incidence rates of approximately 4–5/100,000 per year, ¹ we estimate that approximately 150–200 new CLL cases are diagnosed annually in Bulgaria. Treatment for CLL in Bulgaria typically begins in haematology clinics, where a multidisciplinary haemato-oncology committee evaluates the patient’s clinical status and cytometric, molecular and genetic profile. Local protocols, recently updated in the National Pharmacotherapeutic Guide for Haematology, align with major international treatment guidelines. Although all CLL drugs approved by the European Medicines Agency (EMA) are reimbursed by Bulgaria’s National Health Insurance Fund,^15,16 recent data on the CLL treatment landscape and the economic burden of the disease remain sparse. Therefore, this observational study was conducted to address knowledge gaps regarding the real-world application of FLT in Bulgarian patients with active CLL and to characterise the patient population receiving these treatments.

Materials and methods

Study design and data sources

DESCRIBE was an observational, multicentre, retrospective study based on secondary data collection, conducted in Bulgaria between 1 August 2023 and 16 November 2023. Eight sites participated in this physician-led review of medical charts, collecting data on the demographic, clinical and treatment characteristics of patients with CLL who had active disease and started FLT after 1 January 2022. No follow-up information after FLT initiation was collected. All patients provided written informed consent for study participation and data collection.

Data from all participating centres were consolidated into a single anonymised dataset using an electronic case report form (eCRF). The eCRF was completed by participating physicians or their designated clinical research staff for each eligible patient via a secure, password-protected, web-based data capture system.

The study protocol was approved by the Ethics Committee for Clinical Trials (approval number ЕККИ 0433/17.05.2023). The study was conducted in accordance with the Declaration of Helsinki (1975, as revised in 2024) and the International Society for Pharmacoepidemiology (ISPE) Guidelines for Good Pharmacoepidemiology Practices (GPP). Study outcomes were reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ¹⁷

Study objectives

The primary objective was to describe the demographic and clinical characteristics of patients with active CLL who required FLT in routine clinical practice. Secondary objectives included examining patterns of FLT use in this patient population and assessing correlations between FLT choices and specific patient characteristics.

Study population inclusion and exclusion criteria

Eligibility required participants to be adult patients (≥18 years) with a confirmed diagnosis of CLL and active disease requiring FLT initiation after 1 January 2022, based on the International Workshop on Chronic Lymphocytic Leukemia (iwCLL) guidelines ¹¹ active disease criteria. Patients were also required to provide written informed consent for study data collection and were enrolled consecutively in the order of signing the informed consent form. Exclusion criteria included patients who had started FLT as part of an interventional trial, those who initiated FLT before 1 January 2022 or those enrolled in the study less than 30 days after starting FLT.

Study outcomes

Data collected from medical records included a range of patient-specific variables: demographic information (age and sex), anthropometric measurements (body mass index (BMI)) and medical history, including cardiovascular diseases (CVD) and other comorbidities. Additional variables encompassed concomitant medications at FLT initiation, family history of malignancies and CVD and disease-specific characteristics such as disease duration, Rai/Binet staging, Eastern Cooperative Oncology Group performance status (ECOG PS), laboratory parameters with local normal reference ranges (lower limit of normal (LLN) and upper limit of normal (ULN)) and cytogenetic and molecular abnormalities. The presence of splenomegaly and lymphadenopathy at FLT initiation, as assessed by the treating physician, as well as active disease criteria based on the iwCLL guidelines, ¹¹ were also documented. Treatment-related data included the FLT start date, drug class and specific active agents used.

For secondary outcomes, the following clinical characteristics were considered: age (≥65 years versus <65 years), ECOG PS (0–1 versus ≥2), Rai stage (0–II versus III–IV), Binet stage (A–B versus C), cytogenetic and molecular abnormalities (high-risk, defined by the presence of any of del(17p), del(11q), tumour protein p53 gene (TP53) mutation and/or immunoglobulin heavy chain variable region gene (IGHV)-unmutated, versus no high-risk abnormalities) and comorbidities (comorbidities of interest versus no comorbidities).

Statistical analysis

The target sample size was set at approximately 80–90 patients, based on preliminary feasibility assessments at study sites. This included both newly diagnosed cases with advanced disease and older indolent cases that progressed to aggressive disease requiring FL treatment, regardless of the time of CLL diagnosis.

Descriptive statistics were used to summarise study outcomes and were performed only on observed cases. Categorical variables were reported as frequencies and proportions, and continuous variables as mean with SD and/or median with minimum and maximum, as appropriate. The proportion of missing data was reported for each measured variable.

Analyses were performed on the full analysis set (FAS), which included all eligible patients. The same analyses were conducted on subgroups stratified by the presence of comorbidities (any comorbidity subgroup), high-risk molecular and/or cytogenetic abnormalities (high-risk CLL subgroup) and lymphocyte count (threshold 30 × 10⁹ cells/L). The pre-specified threshold of 30 × 10⁹ cells/L was operationally used in clinical practice as an indicator of tumour burden and is relevant for tumour lysis syndrome (TLS) monitoring during BCL2i initiation. The relationship between clinical characteristics of interest and FLT choice was analysed using regression analysis.

Generalised linear models (GLM) were employed to assess the relationships between patient characteristics (sex, age, BMI, disease duration, comorbidities and high-risk CLL) and disease progression from diagnosis to FLT initiation. Associations between FLT type and patient characteristics were explored using the Pearson chi-square test, Fisher’s exact test and Mann–Whitney–Wilcoxon tests. Statistical significance was set at p < 0.05. Data analyses were performed using SPSS version 26, and logistic regression models were performed in Stata version 17.

Results

Patient and disease characteristics

Overall, 91 patients with a confirmed diagnosis of active CLL requiring FLT initiation were enrolled in the study. Of these, 90 met the study inclusion criteria and were included in the FAS. One patient was excluded from the FAS because enrolment occurred less than 30 days after FLT initiation.

The mean age (SD) of the study population was 62.8 (11.2) years, with over one-quarter of patients (28%) under 60 years of age. The male-to-female ratio was 1.4:1. Table 1 provides a summary of demographic characteristics. Across subgroups, the mean age at CLL diagnosis was generally consistent, except in the subgroup with lymphocyte counts <30 × 10⁹ cells/L, where the mean age at diagnosis was slightly higher than that of the FAS (Table 1). Additionally, the median duration of CLL at enrolment was shorter among patients with high-risk CLL and those with lymphocyte counts <30 × 10⁹ cells/L compared with the FAS (Table 1).

Table 1.

Patient characteristics in the FAS and subgroups.

Characteristics	FAS(N = 90)	Subgroups
		Any comorbidity(N = 58)	High-riskCLL(N = 37)	Lymphocytes
				<30 × 109 cells/L(N = 17)	≥30 × 109 cells/L(N = 72)
Age at CLL diagnosis, years
Mean (SD)	62.8 (11.2)	64.8 (9.77)	63.8 (9.56)	70.5 (7.36)	60.9 (11.2)
Median (min–max)	65 (37–84)	66 (43–80)	62 (43–84)	69 (56–84)	62 (37–80)
<60 years, n (%)	25 (27.8)	16 (27.6)	13 (35.1)	1 (5.9)	29 (40.3)
Male, n (%)	52 (57.8)	34 (58.6)	23 (62.2)	10 (58.8)	42 (58.3)
Duration of disease at the time of enrolment, months
Median (min–max)	19.2 (1.2–206.4)	18 (1.2–206.4)	10.8 (1.2–128.4)	9.6 (1.3–69.6)	1.7 (1.2–206.4)
Duration of disease at the time of FLT start, months
Median (min–max)	6 (0.0–198)	4.8 (0.0–198)	2.4 (0.0–112.8)	–	–
BMI, kg/m2
Mean (SD)	28.0 (5.0)	28.5 (5.44)	28.8 (6.20)	27.1 (6.15)	28.3 (4.66)
Rai/Binet stage at diagnosis, n (%)
Rai stage
0	15 (16.7)	7 (12.1)	3 (8.1)	1 (5.9)	13 (18.1)
I	17 (18.9)	14 (24.1)	6 (16.2)	2 (11.8)	15 (20.8)
II	34 (37.8)	18 (31.0)	15 (40.5)	8 (47.1)	26 (36.1)
III	12 (13.3)	9 (15.5)	6 (16.2)	4 (23.5)	8 (11.1)
IV	12 (13.3)	10 (17.2)	7 (18.9)	2 (11.8)	10 (13.9)
Binet stage
A	30 (33.3)	15 (25.9)	6 (16.2)	3 (17.6)	26 (36.1)
B	41 (45.6)	28 (48.3)	21 (56.8)	9 (52.9)	32 (44.4)
C	19 (21.1)	11 (19.0)	10 (27.0)	5 (29.4)	14 (19.4)
Rai/Binet stage at the time of FLT start, n (%)
Rai stage
0	1 (1.1)	0 (0.0)	0 (0.0)	1 (5.9)	0 (0.0)
I	10 (11.1)	6 (10.3)	2 (5.4)	1 (5.9)	9 (12.5)
II	40 (44.4)	25 (43.1)	16 (43.2)	9 (52.9)	30 (41.7)
III	24 (26.7)	17 (29.3)	11 (29.7)	5 (29.4)	19 (26.4)
IV	15 (16.7)	10 (17.2)	8 (21.6)	1 (5.9)	14 (19.4)
Binet stage
A	10 (11.1)	3 (5.2)	1 (2.7)	2 (11.8)	8 (11.1)
B	57 (63.3)	39 (67.2)	25 (67.6)	12 (70.6)	44 (61.1)
C	23 (25.6)	16 (27.6)	11 (29.7)	3 (17.6)	20 (27.8)
ECOG PS at diagnosis, n (%)
0	37 (41.1)	21 (36.2)	15 (40.5)	3 (17.6)	33 (45.8)
1	34 (37.8)	25 (43.1)	15 (40.5)	6 (35.3)	28 (38.9)
2	17 (18.9)	11 (19.0)	6 (16.2)	8 (47.1)	9 (12.5)
3	1 (1.1)	0 (0.0)	0 (0.0)	0 (0.0)	1 (1.4)
Missing	1 (1.1)	1 (1.7)	1 (2.7)	0 (0.0)	1 (1.4)
ECOG PS at the time of FLT start, n (%)
0	21 (23.3)	11 (19.0)	10 (27.0)	3 (17.6)	18 (25.0)
1	40 (44.4)	28 (48.3)	17 (45.9)	6 (35.3)	33 (45.8)
2	26 (28.9)	19 (32.8)	9 (24.3)	7 (41.2)	19 (26.4)
3	3 (3.3)	0 (0.0)	1 (2.7)	1 (5.9)	2 (2.8)
Laboratory parameters at the time of FLT start, mean (SD)
Lymphocytes*,  ×109/L	86.0 (64.2)	80.9 (55.8)	97.2 (74.6)	15.0 (6.6)	102.8 (60.0)
Platelets*, ×109/L	174.8 (82.9)	164.4 (81.0)	165.8 (83.1)	208.1 (85.8)	167.0 (80.8)
Haemoglobin*, g/L	116.2 (25.7)	112.9 (26.0)	113.8 (26.0)	119.6 (23.8)	115.4 (26.2)
LDH**, U/L	450.0 (337.5)	439.0 (245.5)	504.0 (315.6)	384.5 (118.8)	466.9 (372.6)

*Available for 89 patients in the FAS; **Available for 78 patients in the FAS

BMI: body mass index; CLL: chronic lymphocytic leukaemia; ECOG PS: Eastern Cooperative Oncology Group performance status; FAS: full analysis set; FLT: frontline treatment; LDH: lactate dehydrogenase; min: minimum; max: maximum; NA: not applicable; N: total number of patients; n (%): number (percent) of patients in a specific category.

In total, 58 patients (64%) had at least one comorbidity. The most frequent were CVD (n = 51, 57%), with hypertension being the most common (n = 48, 53%), followed by coronary artery disease/ischaemic heart disease (n = 11, 12.2%), atrial fibrillation (n = 8, 9%), congestive heart failure (n = 5, 6%), rhythm disorders (n = 4, 4%), dyslipidaemia and myocardial infarction (n = 3, 3% each) and peripheral artery disease (n = 1, 1%). Other baseline comorbidities were reported in ≤10% of patients, including chronic pulmonary disease (n = 5, 5.6%), diabetes mellitus (n = 9, 10%), infectious diseases (n = 4, 4.4%), hepatic disorders (n = 2, 2%), renal disorders (n = 4, 4.4%) and other malignancies (all solid tumours; n = 5, 5.6%).

At diagnosis, the largest proportion of patients presented with Rai stage II (38%) or Binet stage B (46%), and most patients (78%) had an ECOG PS of 0–1 (Table 1). Disease progression, as indicated by Rai and Binet staging at FLT initiation, is depicted in Figures S1A and S2B, respectively. Between-subject effects tests suggested that disease duration (p = 0.02) and diabetes (p = 0.036) impacted Rai stage progression, whereas the presence of comorbidities affected Binet stage progression (p = 0.012). The proportion of patients with an ECOG PS of 2 increased from 19% at diagnosis to 29% at FLT initiation (Table 1).

No significant associations were observed between lymphocyte or platelet counts and age, sex, BMI, CVD or high-risk cytogenetic or molecular characteristics at FLT initiation. In subgroups, more than half of the patients had haemoglobin values below the LLN. Patients with comorbidities were more likely to have haemoglobin values <LLN (p = 0.021), based on each site’s laboratory reference ranges. Over half of the patients had platelet counts within the normal range, regardless of subgroup, whereas approximately one-third or more had counts <LLN, with higher proportions observed in the any comorbidity subgroup. No significant associations were found between other laboratory parameters across any of the three subgroups (p > 0.05). The most common concomitant medication was antihypertensive treatment, used by 45 patients (50%). Other treatments included novel oral anticoagulants in five patients (6%) and vitamin K antagonists in three patients (3%). No concomitant treatment was reported for 40 patients (44%).

Molecular genetic testing was performed in fewer than 50% of patients in the FAS, except for del(17p), which was tested in 69% of patients. Del(11q), TP53 and IGHV mutation status were tested in 44%, 21% and 37% of patients, respectively. High-risk del(11q), del(17p) and TP53 mutations were detected in 9%, 13% and 4% of all study patients, corresponding to 20%, 19% and 21% of patients tested for each abnormality, respectively (Figure 1). IGHV-unmutated status was identified in 26% of all study patients or 30% of patients tested for this abnormality. Additional molecular tests were performed in 13% of patients.

Figure 1.

Percentage of patients with cytogenetic or molecular abnormalities in the full analysis set (FAS, N = 90), including cases in which molecular genetic tests were not performed. *Amplification of chromosome 12 (n = 2), del(13q) (n = 4), DLEU deletion in 80% (n = 1), no other cytogenetic abnormalities (n = 4) and trisomy 12 (n = 1). del(11q): chromosome 11q deletion; del(17p): chromosome 17p deletion; CG: cytogenetic; IGHV: immunoglobulin heavy chain variable region; N: total number of patients; TP53: tumour protein p53 gene.

Real-world treatment patterns for active CLL

All patients included in the FAS initiated FLT for active CLL between January 2022 and September 2023. TT was administered in 77% of cases: 46% of patients received BTKi (either as monotherapy or combined with a CD20 antibody), and 31% received BCL2i (either as monotherapy or combined with a CD20 antibody). CIT/chemotherapy (ChT) was given to 23% of patients, with the bendamustine + rituximab combination being the most common (13%) (Figure 2(a)).

Figure 2.

Frontline treatments overall and by class, including monotherapies and combination therapies, in the FAS and relevant subgroups. *included chlorambucil + rituximab in 1 patient, R-CVP in 2 patients and rituximab + cyclophosphamide + vincristine in 1 patient; **included chlorambucil + rituximab in 1 patient and rituximab + cyclophosphamide + vincristine in 1 patient; *** included R-CVP in 1 patient and rituximab + cyclophosphamide + vincristine in 1 patient; ****for the <30 × 10⁹/L subgroup included chlorambucil + rituximab in 1 patient and R-CVP in 1 patient, while for the ≥30 × 10⁹/L subgroup included R-CVP in 1 patient and rituximab + cyclophosphamide + vincristine in 1 patient. BCL2i: B-cell lymphoma-2 inhibitor; BTKi: Bruton tyrosine kinase inhibitor; CIT: chemoimmunotherapy; ChT, chemotherapy; N: total number of patients; n: number of patients in a specific category; R-CVP: rituximab + cyclophosphamide + vincristine +prednisolone; FAS: full analysis set.

In subgroups, the most commonly used treatments were BTKi (62% in the high-risk subgroup and 46%–47% in subgroups stratified by lymphocyte count) and BCL2i (47%) in the subgroup with any comorbidity (Figure 2(b) to (d)).

The distribution of CLL treatment classes varied significantly according to ECOG PS at CLL diagnosis (p < 0.001) and FLT initiation (p = 0.005), with a strong association between ECOG PS 0–1 and TT use as well as ECOG PS ≥2 and CIT/ChT use (Table S1A). A medical history of CVD was negatively associated with TT administration (p = 0.014) (Table S1B). TT selection was significantly related to both ECOG PS at diagnosis (p = 0.002) and at FLT start (p = 0.03) (Table S1A). First-generation BTKi monotherapy use was significantly associated with Rai stage 0–II at FLT initiation (p = 0.034) (Table S1C). CIT selection was significantly associated with ECOG PS ≥2 at CLL diagnosis (p < 0.001), FLT start (p = 0.011) and the presence of comorbidities (p = 0.047) (Table S1A and S1D). ECOG PS 2 at CLL diagnosis increased the likelihood of using specific CIT regimens, i.e. bendamustine + rituximab, chlorambucil + obinutuzumab or fludarabine +cyclophosphamide + rituximab (p < 0.001). Patients with high-risk CLL were more likely to receive second-generation BTKi (odds ratio (OR) = 5.59, p < 0.001), with each additional high-risk trait further increasing the odds of its use (OR = 2.35, p < 0.01). A history of CVD and age were other factors that significantly influenced CIT selection (OR = 0.18, p < 0.05 and OR = 1.07, p < 0.05, respectively). These results should be interpreted with caution due to the low number of patients.

Discussion

This observational study provides valuable insights into the clinical characteristics and FLT patterns for patients with active CLL in real-world settings in Bulgaria after 1 January 2022. DESCRIBE represents one of the first noninterventional studies in this patient population in Bulgaria, offering real-world evidence on the profile of CLL patients and the contemporary FLT landscape (2022–2023).

The cohort included 90 patients, with a median age of 65 years at diagnosis and 67 years at FLT initiation. More than half of the patients (58%) were male, and nearly two-thirds had at least one comorbidity (64%), predominantly CVD (57%). The median time from CLL diagnosis to FLT initiation was 6 months. These findings indicate that the CLL patient population in Bulgaria is slightly younger than that reported in the United States and Western Europe, where the median age at diagnosis is typically 70–72 years.^1,5 Furthermore, the male-to-female ratio in our cohort was 1.4:1, which is lower than the ratios commonly observed in western countries (1.7:1–1.9:1).^5,18 A younger age at FLT initiation was also observed in patients with high-risk CLL, consistent with previous reports indicating that younger CLL patients (≤55 years) tend to have more aggressive disease and require earlier treatment initiation. ¹⁹

The high incidence of CVD in this cohort aligns with findings from other real-world studies, reflecting the elderly nature of the CLL population and the impact of comorbidities on disease prognosis and treatment decisions.^20–25 In this context, comorbidities can influence both prognosis and treatment selection.^21,23–25 For example, based on actual patterns in our cohort, patients with CVD on direct anticoagulants or vitamin K antagonists were recommended to avoid BTKi when possible due to bleeding risk, and BCL2i ± CD20 antibody was preferred. For frail patients (ECOG ≥2), CIT was selected because of fixed treatment duration and easier toxicity monitoring. In patients with high tumour burden and concomitant CVD, BTKi was preferred over BCL2i if the risk of TLS was high and comorbidities limited inpatient ramp-up monitoring. We also observed that patients with comorbidities were more likely to have haemoglobin values below the LLN (p = 0.021), which may reflect anaemia of chronic disease, antithrombotic therapy or impaired renal function. These findings highlight that standardised assessments of comorbidities and functional status are essential for guiding treatment selection and overall disease management; however, such assessments remain inconsistent across CLL practices. ²⁰

The advent of TT has significantly reshaped the treatment landscape of CLL, resulting in notable improvements in survival outcomes.^{1,5,10,11,18,26} Treatment initiation is typically guided by the presence of symptoms and the risk of disease progression.^1,5,10,18 Although many patients in our dataset had Rai 0–II/Binet B disease, all treatment initiations met the iwCLL ¹¹ criteria for active disease, including progressive cytopenias, symptomatic lymphadenopathy or splenomegaly, rapid lymphocyte doubling time or disease-related symptoms. The most common indications were marrow compromise and symptomatic lymphadenopathy. In recent years, the number of targeted agents approved for CLL has increased. The selection of the most appropriate therapeutic regimen is now strongly influenced by factors such as advanced age and poor performance status, which are closely associated with an increased risk of toxicity and treatment intolerance. ⁷ Consequently, treatment decisions in CLL are multifaceted and depend on several key considerations, including age, performance status and comorbidities, particularly in relation to the potential toxicities of novel therapies. Additionally, the presence or absence of high-risk genetic or molecular abnormalities, such as TP53 mutations, del(17p) and IGHV mutational status, plays a pivotal role in shaping treatment choices. Other important factors include the availability or reimbursement of specific therapies as well as patient preferences.^27,28 In the current CLL treatment landscape, testing for specific genetic and molecular markers is critical for both disease prognosis and management. Previous studies have demonstrated that patients with del(17p) and, to a lesser extent, del(11q) who were treated with CIT had poorer outcomes. ²⁹ Similarly, patients with del(17p) and/or TP53 mutations exhibited inferior responses to CIT. However, the introduction of TT has led to significantly improved outcomes for these patients. ⁵ Consistent with findings from a large real-world evidence study conducted in the US, ²⁹ in which cytogenetic testing rates were lower than expected, similar trends were observed in the DESCRIBE study. Specifically, del(17p) was tested in 69% of patients, while del(11q), TP53 and IGHV mutation status were assessed in 44%, 21% and 37% of patients, respectively. In Bulgaria, the costs of onco-haematological diseases are covered by the NHIF through distinct pathways for drugs, hospitalisations and clinical testing. CLL care is centralised in eight tertiary haematology centres, where a multidisciplinary tumour board determines treatment. Cytogenetic testing is available at all centres, whereas IGHV and TP53 sequencing availability varies and may require collaboration with external laboratories. Higher co-payment costs for diagnostic tests may partly explain the lower testing rates observed in this study. ³⁰ Differences in testing rates also reflect the sequential implementation of testing technologies in Bulgarian centres, reimbursement restrictions and clinical urgency that occasionally necessitates treatment initiation before molecular results become available. Currently, testing for IGHV and TP53 mutation status is not reimbursed by the NHIF but can be conducted with support from pharmaceutical companies. Fluorescence in situ hybridisation analysis is available and performed in the majority of patients, while traditional cytogenetics, although available, are not routinely performed for all patients. In the era of precision medicine, improving testing rates for high-risk molecular and genetic markers is essential to enhance outcomes for patients with CLL.

The results of the DESCRIBE study suggest that the treatment landscape for CLL in Bulgaria is now consistent with broader global trends,^5,27 with TT involving BTK and BCL2i emerging as the predominant treatment options for patients with active disease (77% of cases in this study). In Bulgaria, all TT medications are fully reimbursed. Decisions regarding reimbursement, pricing and disease coverage are made by the National Council for Prices and Reimbursement, ¹⁶ and costs are covered by the NHIF. However, a notable gap exists between EMA approval and local reimbursement, with a median delay of 611 days for general medications and 685 days for oncology treatments in Bulgaria. ³¹ This delay significantly affects treatment decisions and is compounded by the fact that access to care for patients with CLL is limited to a small number of specialised centres, predominantly located in major cities.

Real-world evidence is important for clinicians and payers, as it helps establish realistic expectations for outcomes with novel treatments. ³² At present, CIT is no longer considered standard FLT according to international guidelines.^1,10,11 Analyses of large databases from the US illustrate how TT has replaced CIT over time. For example, the informCLL registry, which included 854 patients requiring FLT between 2015 and 2019, showed similar proportions of first-generation BTKi and CIT use at that time (46% versus 43%). ³³ Analysis of the Optum database demonstrated that the difference in FLT distribution widened over time, from 2014 to 2017 (n = 2585; 42% TT vs 30% CIT) to 2018–2021, when FLT was dominated by TT (n = 2641; 54% TT vs 16% CIT). ³⁴ In Western Europe, a study from Finland indicated a shift from conventional CIT to TT between 2014 and 2019, ³⁵ while in Germany, in a cohort of 2983 patients with confirmed CLL, CIT was the predominant FLT until 2019, with TT use gradually increasing from 3% in 2015 to 77% in 2022. ³⁶ Similarly, in Spain, the SRealCLL study showed rapidly increasing use of ibrutinib after market authorisation, with a corresponding decline in CIT/CT use. ³⁷ At the Central and Eastern Europe (CEE) level, similar trends are presumed; however, recent published data on real-world FLT patterns across countries in this region are lacking. In our study, in rare cases where CIT was used in patients later identified as high risk, treatment was initiated before molecular results became available or access to TT was delayed. One frail elderly patient (ECOG ≥2, multiple comorbidities) received chlorambucil monotherapy as an individualised decision based on treatment tolerance. The use of CIT/CT in our cohort likely reflects physician familiarity, historical practice patterns, delayed access to novel agents and possible concerns regarding cardiovascular toxicity, rather than evidence-based superiority. These findings underscore the need for ongoing education and system-level optimisation to better align clinical practice with current standards. Additional real-world evidence evaluating subsequent lines of therapy, survival and economic outcomes is needed to comprehensively understand the burden of CLL in Bulgaria and other countries in the region.

Based on their mechanism of action, drug interactions between novel TT and medications used to treat co-existing conditions and/or de novo toxicities, mainly within the spectrum of CV and infectious diseases, are highly relevant in the day-to-day clinical care of patients with CLL. A thorough history of comorbidities, treatments in use and ongoing re-evaluations within cross-functional teams are essential for optimal outcomes. ³⁸ An important topic for haematologists, particularly in light of the potential CV complications associated with the long-term use of BTKi therapy, is understanding CV risk through proactive CV surveillance and appropriate management of cardiotoxicities without unnecessary interruptions to CLL treatment. In this context, a new concept in the field of cardio-oncology has been introduced, known as ‘permissive cardiotoxicity’, ³⁹ which refers to maintaining effective oncological treatment when cardiotoxicity is detected, without discontinuation. This approach involves multidisciplinary collaboration between haemato-oncologists and cardiologists. In the CLL population, the true rate of treatment-related cardiotoxicities encountered in real-world settings is not well established. Furthermore, a population-based study conducted in Sweden in patients with CLL before starting any BTKi therapy showed that one-third had baseline CVD at diagnosis or at the start of anti-leukaemic treatment. ²² In our study, more than half of the patients (57%) already had CVD at the time of FLT initiation, which is consistent with the advanced median age and literature reports of 33%–50% CVD prevalence at CLL diagnosis. ²² Identifying patients at high CV risk or with established CVD would help clinicians not only in the initial treatment decision but also in implementing appropriate CVD surveillance after FLT initiation to improve treatment tolerability and overall outcomes without discontinuing anticancer treatment under the ‘permissive cardiotoxicity’ concept.^39–41

Collecting data in a consistent, standardised manner across multiple clinical centres enabled the generation of robust real-world evidence in Bulgaria. However, due to its observational nature, inclusion only of patients alive at study start, and retrospective data collection from medical charts as recorded in routine clinical practice, the findings of the DESCRIBE study are subject to selection bias, limitations in data availability, incomplete molecular profiling across the cohort and variability in treatment practices, all of which impact the generalisability of our results. In addition, because of the small sample size, the findings from our study should be interpreted with caution. In DESCRIBE, no post-FLT data were collected; therefore, no results on treatment efficacy and safety are available. Although no structured CV monitoring was in place in routine practice at the time of study conduct, the study did not collect information in this regard. Nevertheless, the characterisation of FLT patterns in routine clinical practice is of significant interest to the haematology community, particularly given the lack of recent, systematically collected data on treatment trends and practices.

Conclusion

This observational study characterises the clinical profile and FLT patterns employed in routine practice after 2022 among a cohort of 90 patients with active CLL in Bulgaria. In the era of precision medicine, increasing the rate of testing for specific high-risk molecular and genetic markers is crucial to improving patient outcomes. Our findings indicate a shift towards the adoption of novel therapies, with BTK and BCL2i now predominating the FLT landscape. Given the established efficacy of targeted agents, chemo-immunotherapy should be discouraged in accordance with current international guidelines. Future analyses examining treatment patterns in greater detail—including subsequent lines of therapy, survival outcomes and economic impacts—are necessary to fully characterise the multifaceted burden of CLL.

Author contributions

VG, AR, IM, MEE, MD, TY, VSP, SN, EH, LGK, MPL, IN and ZGP contributed to data acquisition, analysis, data interpretation and writing–review and editing. VG, IM, VSP and ZGP contributed to writing–original draft. SK contributed to the conceptualisation of the study. AN, AO and SK contributed to project administration, analysis, data interpretation and writing of the original draft. All authors reviewed and commented on each version of the manuscript and read and approved the final version.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request. Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca’s data sharing policy described at https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure.

Declaration of conflicting interests

M.E. Efraim declares support for the present manuscript for medical writing and article processing charges from AstraZeneca. S. Krasteva declares support for the present manuscript for medical writing and article processing charges from AstraZeneca. At the time of study conduct, S. Krasteva was employed by AstraZeneca and is currently employed by Astellas Pharma. L. Gercheva-Kyuchukova declares support for the present manuscript for medical writing and article processing charges from AstraZeneca as well as payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from AbbVie, Johnson & Johnson, Novartis, Roche and Swixx. V. Graklanov declares support for the present manuscript for medical writing and article processing charges from AstraZeneca; payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from AstraZeneca, Astellas, Janssen, Novartis, Pfizer, Roche, Swixx and Takeda; and support for attending meetings and/or travel from AstraZeneca, Janssen and Roche. Z. Grudeva-Popova declares support for the present manuscript for medical writing and article processing charges from AstraZeneca as well as payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from AstraZeneca, Astellas, Janssen, Novartis, Pfizer, Roche, Swixx and Takeda as well as support for attending meetings and/or travel from AstraZeneca, Astellas, Janssen, Novartis, Pfizer, Roche, Swixx and Takeda. E. Hadjiev declares support for the present manuscript for medical writing and article processing charges from AstraZeneca; payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from Amgen, Roche and Takeda; support for attending meetings and/or travel from AbbVie and Johnson & Johnson; and participation on a Data Safety Monitoring Board or Advisory Board from AbbVie and Roche. I. Mitcheva declares support for the present manuscript for medical writing and article processing charges from AstraZeneca; payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from Amgen, Astellas, AstraZeneca, Bristol Myers Squibb, Johnson & Johnson, Novartis, Pfizer, Roche and SOBI; payment for expert testimony from Astellas, AstraZeneca, Bristol Myers Squibb, Janssen and Novartis; support for attending meetings and/or travel from Roche; and participation on a Data Safety Monitoring Board or Advisory Board from Astellas, AstraZeneca, Bristol Myers Squibb, Janssen and Novartis. A. Nenov declares support for the present manuscript for medical writing and article processing charges from AstraZeneca and is an AstraZeneca employee. A. Ostojić declares support for the present manuscript for medical writing and article processing charges from AstraZeneca and is an AstraZeneca employee. A. Radinoff declares support for the present manuscript for medical writing and article processing charges from AstraZeneca; payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events from Roche and Swixx; support for attending meetings and/or travel from Roche; and participation on a Data Safety Monitoring Board or Advisory Board from Astellas, AstraZeneca, Johnson & Johnson, Novartis, Pfizer, Roche, Servier, SOBI and Swixx. M. Donchev, S. Naneva, I. Nikolov, M. Petrova-Latkova, V. Slavcheva-Popova and T. Yankova declare support for the present manuscript for medical writing and article processing charges from AstraZeneca.

Ethical considerations

The study was approved by the Ethics Committee for Clinical Trials (approval number ЕККИ 0433) on 17 May 2023.

Funding

This study was sponsored by AstraZeneca. AstraZeneca contributed to the study design and participated in data collection, analysis, interpretation, writing, reviewing and approval of the final version. No honoraria or payments were made for authorship.