Toward the future management of patients with CML and Ph + ALL: real-world safety insights from dasatinib pharmacovigilance

Background: Dasatinib, a second-generation BCR-ABL1 tyrosine kinase inhibitor, has transformed treatment for chronic myeloid leukemia and Philadelphia chromosome–positive acute lymphoblastic leukemia. However, its broad kinase inhibition leads to distinct adverse events (AEs), requiring systematic post-marketing surveillance.

Objective: To evaluate dasatinib-associated AEs using the United States Food and Drug Administration Adverse Event Reporting System (FAERS) and identify potential safety signals.

Methods: We extracted FAERS reports (Q1 2004–Q4 2024) listing dasatinib as the primary suspect drug, submitted by physicians or pharmacists. After data cleaning, AEs were coded using MedDRA terminology. Signal detection was performed with four disproportionality methods (ROR, PRR, BCPNN, MGPS). Time-to-onset was analyzed using Weibull models, with subgroup analyses by demographic and clinical characteristics.

Results: Among 7,213 cases, respiratory disorders were the most prominent signal (ROR = 2.74). A total of 1,951 significant signals were identified, 113 confirmed by all algorithms. Strongest disproportionality signals included blast cell proliferation (ROR = 518.94), primary effusion lymphoma (ROR = 181.38), and allogeneic bone marrow transplantation therapy (ROR = 171.48). Common AEs were pleural effusion (n = 828, ROR = 35.87), hepatotoxicity (n = 262, ROR = 29.17), and fluid retention (n = 129, ROR = 14.49). Weibull analysis showed an early failure pattern (β = 0.64), with 25.6% of AEs within 30 days and 28.2% after 360 days. Subgroup analysis revealed pleural effusion across all demographics, while hematologic signals predominated in adults and elderly patients.

Conclusion: This large FAERS analysis characterized known respiratory complications of dasatinib, particularly pleural effusion, and identified several rare but potentially relevant safety signals hepatotoxicity, cardiovascular events, and novel AEs such as alveolar proteinosis and cytomegalovirus-related complications. The early failure pattern underscores the need for intensive monitoring during initiation and continued vigilance during long-term therapy. These findings provide real-world, hypothesis-generating evidence to optimize safety management and inform future clinical investigations, and should be interpreted with caution in view of the inherent limitations of spontaneous reporting data.

Introduction

Chronic myeloid leukemia (CML) and Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph + ALL) are hematological malignancies driven by Breakpoint Cluster Region-Abelson murine leukemia viral oncogene homolog 1 (BCR-ABL1) fusion oncoprotein, which constitutively activates tyrosine kinase signaling and promotes malignant proliferation and survival (1, 2). CML is characterized by an initial chronic phase that, if untreated, progresses to an accelerated phase and a blast phase, both of which are associated with poor prognosis (3). Ph + ALL represents a subset of acute lymphoblastic leukemia with an adverse prognosis compared to BCR-ABL1-negative disease, due to its aggressive clinical course and high relapse rate despite intensive chemotherapy (4). The advent of molecular-targeted therapies has transformed the management of these disorders, improving remission rates and overall survival (5).

Dasatinib is a second-generation BCR-ABL1 tyrosine kinase inhibitor approved for treating chronic, accelerated, and blast-phase CML as well as Ph + ALL, particularly in patients who are resistant or intolerant to imatinib (6). In pivotal clinical trials, dasatinib achieved higher rates of complete cytogenetic response and major molecular response compared to imatinib in newly diagnosed chronic-phase CML, and it also induced durable remission in imatinib-resistant disease (7). However, the kinase inhibitory profile of dasatinib extends beyond BCR-ABL1 to include Src family kinases, c-KIT, platelet-derived growth factor receptor (PDGFR), and EphA2 (an ephrin-A family receptor), contributing to a distinct spectrum of AEs (8, 9). Common AEs observed in clinical studies include myelosuppression (neutropenia and thrombocytopenia), fluid retention (pleural effusion and peripheral edema), gastrointestinal disturbances, and dermatological reactions (10).

Less common, but serious toxicities include pulmonary arterial hypertension, cardiac dysfunction, and hemorrhagic events (the latter is linked, in part, to dasatinib-associated platelet dysfunction) (11, 12). These toxicities have prompted dose modifications, treatment interruptions, and, in rare cases, discontinuation, underscoring the need for systematic pharmacovigilance to characterize the real-world safety profiles (10).

The United States Food and Drug Administration Adverse Event Reporting System (FAERS) is a public database that collates spontaneous AE reports submitted by healthcare professionals, consumers, and manufacturers worldwide. FAERS supports post-marketing surveillance by enabling the detection of safety signals through disproportionality analysis and temporal trend assessment (13). Although FAERS is subject to underreporting, reporting biases, and lack of denominator data, it remains a cornerstone for hypothesis generation and signal detection, complementing clinical trial data and observational studies.

In this study, we used the FAERS database to systematically evaluate AEs reported for dasatinib, distinguishing reports submitted by internal medicine physicians from those submitted by pharmacists. We aimed to identify and characterize the spectrum of dasatinib-associated AEs in routine clinical practice, thereby informing risk-mitigation strategies and optimizing patient management.

Materials and methods

Data sources and processing

The quarterly FAERS data were obtained from the FDA website. Data cleaning and integration were conducted using R software (version 4.3.1). Reports were retained if the preferred term (PT) count was ≥3. Duplicate entries were removed; for reports with multiple AEs under the same CASE_ID, only the most recent FDA date (FDA_DT) and the higher primary_ID were retained. Records lacking essential information or detailed drug data were excluded. Reports without any coded adverse event term (exposure-only records) were also excluded before signal detection.

Adverse events were coded using the Medical Dictionary for Regulatory Activities to standardize the System Organ Class (SOC) and PT terminology. Reports between Q1 2004 and Q4 2024 were extracted and restricted to those listing dasatinib as a primary suspected drug. To improve reliability, only the cases submitted by physicians or pharmacists were included. The initial dataset contained 22,375,298 reports. After removing 3,788,074 duplicates, 7,213 unique cases were identified (Figure 1). Key demographic and clinical variables, including PRIMARYID, AGE, and GENDER scores, were extracted.

FIGURE 1

Figure 1. The flowchart of the entire study showing the selection of FAERS reports listing dasatinib as the primary suspect drug (Q1 2004–Q4 2024) and reported by physicians or pharmacists.

Signal detection

A 2 × 2 contingency table was constructed to compare the observed frequencies of adverse events between the exposed and non-exposed groups (Supplementary Table 1). Disproportionality analyses were performed using four established methods: reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS). ROR is computationally simple and well suited to large-scale screening, PRR offers an intuitive assessment of disproportionality, BCPNN reduces false positives in small or rare-event datasets, and MGPS incorporates Bayesian shrinkage to address sparse data and multiple testing, making it particularly robust for long-term pharmacovigilance (14) (Supplementary Table 2). For primary signal detection, we predefined the following thresholds based on the 2 × 2 contingency table: for ROR and PRR, at least three reports (a ≥ 3) with the lower limit of the 95% confidence interval greater than 1; for BCPNN, a positive expectation of the information component [0 < E(IC)]; and for MGPS, a lower 95% confidence limit of the empirical Bayes geometric mean greater than 2 with at least one report (EBGM_05 > 2, a > 0) (Supplementary Table 2). Drug–event pairs that met the predefined thresholds of all four algorithms were included in the analysis. Time-to-onset (TTO) was calculated as TTO = AE onset date – dasatinib initiation date. Reports with erroneous entries (such as EVENT_DT earlier than START_DT) were excluded. Median values were used to summarize the timing of onset. A parametric Weibull model was fitted to the TTO distribution to characterize the overall time pattern of reporting; because FAERS does not provide consistent information on treatment line, cumulative exposure, or dose modifications, this analysis was intended to describe reporting dynamics rather than to define specific early- versus late-treatment phases. In addition, sensitivity analyses were performed by recalculating disproportionality metrics stratified by reporter role (physicians vs. pharmacists) and by the three main reporting countries (United States, Japan, and France).

Results

Baseline characteristics

Among the 7,213 included cases (Table 1), female patients accounted for 3,081 (42.7%) and male patients for 3,095 (42.9%). Weight data were largely missing (90.4%, n = 6,522). Among those with available information, 7.4% (n = 533) weighed 50–100 kg, 0.8% (n = 61) weighed < 50 kg, and 1.3% (n = 97) weighed > 100 kg. Patients aged 18–64.9 years represented the largest group (36.5%, n = 2,634), followed by those aged 65–85 years (19.9%, n = 1,436). Children (<18 years) accounted for 3.9% (n = 281), while patients > 85 years comprised 1.2% (n = 84). Serious cases accounted for 53.3% (n = 3,846), whereas non-serious cases comprised 46.7% (n = 3,367). Fatal outcomes were reported in 622 cases (8.6%), with 91.4% (n = 6,591) surviving. Most reports originated from the United States (71.1%, n = 5,126), followed by Japan (7.2%, n = 519) and France (3.1%, n = 226). A temporal trend showed peak reporting in 2015 (n = 993), with a subsequent decline in later years (Figure 2).

TABLE 1

Table 1. Basic information on AE reports for dasatinib.

FIGURE 2

Figure 2. Number of dasatinib-associated AE reports per year.

Signal detection findings

Using ROR, PRR, IC, and EBGM methods, dasatinib-related AEs were distributed across multiple SOCs. As shown in Table 2, the most frequently reported were general disorders and administration site conditions (n = 2,404; ROR 0.98; PRR 0.98; IC −0.03; EBGM 0.98), respiratory, thoracic and mediastinal disorders (n = 2,080; ROR 2.74; PRR 2.52; IC 1.33; EBGM 2.51), gastrointestinal disorders (n = 1,808; ROR 1.41; PRR 1.37; IC 0.45; EBGM 1.37) and nervous system disorders (n = 999; ROR 0.74; PRR 0.76; IC −0.40; EBGM 0.76). Respiratory system signals (ROR = 2.74) were the most prominent, warranting clinical attention. Although general disorders accounted for the highest number of reports, their disproportionality indices were not significant, suggesting non-specific associations rather than drug-specific risks.

TABLE 2

Table 2. Disproportionality analysis of dasatinib-related AEs by SOC.

At the PT level, 1,951 dasatinib-related AE signals were identified, of which 113 signals were validated by all four algorithms. The detected signals encompassed a broad range of clinical manifestations, predominantly involving hematologic malignancy progression, pleural/pulmonary complications, transplantation outcomes, infectious events, and cardiovascular/metabolic effects. The three strongest signals by ROR ranking were: blast cell proliferation (n = 11; ROR 518.94; PRR 518.59; IC 8.39; EBGM 334.93), primary effusion lymphoma (n = 5; ROR 181.38; PRR 181.33; IC 7.25; EBGM 152.24) and allogeneic bone marrow transplantation therapy (n = 4; ROR 171.48; PRR 171.43; IC 7.18; EBGM 145.21). By absolute case counts, the most frequently reported PTs were: pleural effusion (n = 828; ROR 35.87; PRR 34.08; IC 5.04; EBGM 32.92), hepatotoxicity (n = 262; ROR 29.17; PRR 28.72; IC 4.8; EBGM 27.9) and fluid retention (n = 129; ROR 14.49; PRR 14.38; IC 3.83; EBGM 14.18).

In addition to these core adverse events, the analysis also identified several signals related to serosal effusions and respiratory system disorders, such as polyserositis (ROR = 21.34), chylothorax (ROR = 169.55), hydrothorax (ROR = 11.68), and alveolar proteinosis (ROR = 13.22), indicating the broad impact of dasatinib on the pleural–pulmonary environment. Furthermore, AEs associated with infections and immune function were observed, including cytomegalovirus enterocolitis (ROR = 16.07), cytomegalovirus colitis (ROR = 10.98), and neutrophilic dermatosis (23.88), some of which showed high signal strength and warrant clinical attention. Notably, this study not only confirmed well-documented AEs already listed on the dasatinib product label–such as pleural effusion (ROR = 35.87), fluid retention (ROR = 14.49), and hepatotoxicity (ROR = 29.17)–but also revealed more specific or uncommon manifestations, including skin hypopigmentation (ROR = 12.70) and skin depigmentation (ROR = 11.03). Beyond the known risks of myelosuppression and transplantation-related complications, the analysis also identified emerging clinically relevant events, such as hydrops foetalis (ROR = 14.43), peripheral artery stenosis (ROR = 16.61), and ejection fraction abnormal (ROR = 10.28). After excluding PTs attributable to underlying disease progression or treatment-related procedures, several rare but potentially important signals (e.g., alveolar proteinosis, drug tolerance decreased) were identified (Table 3). These findings deserve further validation through clinical and mechanistic research.

TABLE 3

Table 3. Disproportionality analysis of dasatinib-related AEs by PT.

Subgroup analyses

In subgroup analyses stratified by gender, age, and body weight, pleural effusion consistently emerged as the most prominent signal across all categories. Hematologic and cytogenetic events such as blast cell proliferation and Philadelphia chromosome positive were predominantly enriched among adults, elderly patients, and male populations. In contrast, children and low-weight patients exhibited some distinctive signals, including hemorrhagic enterocolitis, prescribed underdose, and gastrointestinal hemorrhage. Patients in the medium- and high-weight groups more frequently reported cardiopulmonary events, such as pericardial effusion, pulmonary hypertension, and pulmonary edema. Overall, these findings suggest that while fluid retention events (e.g., pleural effusion) are common across all populations, the spectrum of dasatinib-related adverse events varies according to demographic and clinical characteristics; however, these apparent differences may also reflect variation in prescribing patterns and reporting behavior between subgroups (Supplementary Tables 3–11).

TTO analysis

A total of 1,500 reports documented the onset time of AEs and were included in the analysis. The results showed that 28.2% (n = 423) of AEs occurred after 360 days of treatment initiation. The distribution across other time intervals was as follows 0–30 days (25.6%, n = 384), 31–60 days (10.3%, n = 155), 61–90 days (7.3%, n = 110), 91–120 days (5.2%, n = 78), 121–150 days (3.9%, n = 59), 151–180 days (4.3%, n = 64), 181–360 days (15.1%, n = 227) (Figure 3). The mean time to onset for dasatinib-related AEs was 346.1 days, with a median onset time of 131.5 days and an interquartile range (IQR) of 29–413 days (Table 4).

FIGURE 3

Figure 3. Distribution of the time to onset (TTO) of dasatinib-associated AEs.

TABLE 4

Table 4. Time to onset analysis (days).

A Weibull distribution model was fitted to characterize the time-to-onset pattern. The estimated scale parameter (α) was 244.65 (95% CI: 224.16–265.15), and the shape parameter (β) was 0.64 (95% CI: 0.61–0.66). Since the 95% CI of β was entirely below 1, the time-to-onset of dasatinib-related AEs was classified as an early failure type, indicating that these events were more likely to occur during the early phase of treatment (Table 5). Because information on treatment line, cumulative exposure and dose modification is not captured in FAERS, this Weibull model should be regarded as a descriptive summary of reporting patterns rather than a formal survival analysis.

TABLE 5

Table 5. Weibull parameters for AEs onset.

Sensitivity analyses

To explore potential reporting bias, we performed sensitivity analyses stratified by reporter type (physicians vs. pharmacists) and by the three main reporting countries (the United States, Japan and France). Across all strata, pleural effusion remained the most prominent dasatinib-associated event, together with persistent signals for pulmonary edema, fluid retention, pericardial effusion and pulmonary hypertension. Hepatotoxicity showed a particularly strong signal in physician-reported and U.S. cases, whereas Japanese and French reports highlighted several rare but high-magnitude serosal and respiratory events, such as chylothorax and pulmonary arterial hypertension, albeit based on few cases. Overall, these findings indicate that the core pulmonary and hepatobiliary safety profile of dasatinib is robust to differences in reporter type and geographic origin (Supplementary Tables 12–16).

Discussion

Major findings and clinical implications

This comprehensive pharmacovigilance study of 7,213 dasatinib-related AEs from the FAERS database identified several critical safety signals that warrant clinical attention and mechanistic exploration. Our analysis identified 113 statistically significant PTs using rigorous four-algorithm screening, with respiratory system disorders emerging as the most prominent safety concern (ROR = 2.74). These findings not only confirm well-established adverse effects, but also highlight previously underreported safety signals that may have significant clinical implications for optimizing dasatinib therapy.

Pleural effusion was the most frequently reported AE (n = 828, ROR = 35.87), which aligns with established clinical knowledge and reinforces the critical importance of pulmonary monitoring during dasatinib therapy (14–16), with clinicians keeping a low threshold for evaluating new or worsening dyspnea, cough, or unexplained weight gain. However, the identification of novel signals, such as blast cell proliferation (ROR = 518.94), primary effusion lymphoma (ROR = 181.38), and allogeneic bone marrow transplantation therapy (ROR = 171.43), should be interpreted with caution. These findings require careful clinical interpretation because they may represent either true drug-related toxicities or confounding by indication, given the underlying malignant conditions being treated (17, 18).

Overall, these patterns support routine baseline and periodic assessment of respiratory symptoms, liver function, and cardiovascular status, especially in patients with pre-existing cardiopulmonary or hepatic comorbidities.

Temporal analysis, revealing an early failure pattern (Weibull β = 0.64, 95% CI: 0.61–0.66), provides crucial insights into clinical monitoring strategies. With 25.6% of AEs occurring within the first 30 days and a significant proportion (28.2%) occurring after more than 360 days, a bimodal surveillance approach is necessary. Early onset events likely reflect acute toxicities related to initial drug exposure, whereas late-onset events may indicate cumulative, dose-dependent effects or long-term off-target consequences (19–21).

To facilitate clinical interpretation, key safety signals and their main monitoring focus are briefly summarized in Table 6; these points are intended as pragmatic reminders rather than formal management recommendations.

TABLE 6

Table 6. Key dasatinib-related safety signals and main monitoring focus.

Mechanistic insights into dasatinib-related AEs

Respiratory and pleural complications

The predominance of respiratory system disorders in our analysis reflects the complex interaction between dasatinib and pulmonary physiology. Pleural effusion, the most commonly reported AE, results from the inhibition of PDGFR and Src family kinases, leading to increased vascular permeability and lymphatic drainage abnormalities (22, 23). The mechanism involves disruption of endothelial barrier function through interference with PDGFR-mediated signaling pathways, which are crucial for maintaining vascular integrity. Additionally, the off-target effects of dasatinib on lymphatic drainage may contribute to fluid accumulation in serous cavities (24, 25).

Identifying chylothorax, hydrothorax, and polyserositis signals suggests a broader serosal involvement beyond simple pleural effusion. Despite being rare, chylothorax represents a more severe manifestation involving lymphatic system disruption, potentially through the effects of dasatinib on lymphatic endothelial cells or thoracic duct integrity (11). The mechanisms underlying these complications remain incompletely understood but may involve alterations in lymphangiogenesis or lymphatic vessel permeability via Src kinase pathway modulation (26).

Alveolar proteinosis, another notable respiratory signal in our analysis, suggests a potential interference with alveolar macrophage function or surfactant metabolism. This effect could result from the impact of dasatinib on macrophage polarization and clearance functions, given the influence of the drug on immune cell signaling pathways (27, 28). The clinical significance of this finding requires further investigation, as alveolar proteinosis can lead to progressive respiratory failure if left unrecognized.

Hepatotoxicity mechanisms

Hepatotoxicity emerged as the second most frequent AE by absolute count (n = 262, ROR = 29.17), reflecting multiple potential mechanisms of liver injury. Dasatinib-induced hepatotoxicity appears to involve disruption of bile acid homeostasis through inhibition of the bile salt export pump (BSEP), leading to intracellular bile acid accumulation and hepatocyte toxicity (29, 30). This mechanism is supported by studies demonstrating competitive inhibition of BSEP by dasatinib, with further elevation of serum biomarkers for hepatotoxicity in 25%–50% of patients (31).

Moreover, dasatinib may induce oxidative stress and mitochondrial dysfunction, thereby triggering autophagy as a protective mechanism. Research has demonstrated that dasatinib activates p38 signaling pathways, which modulate oxidative stress-related liver injury and autophagy (32, 33). The activation of autophagy appears to serve as a cellular defense mechanism, as autophagy inhibition can exacerbate dasatinib-induced liver failure (34). Furthermore, dasatinib treatment has been associated with increased cytochrome P450 7A1 (CYP7A1) expression, leading to enhanced bile acid synthesis and potential hepatotoxic accumulation (35).

The involvement of immune-mediated mechanisms in dasatinib hepatotoxicity has been suggested by studies demonstrating significant lymphocyte infiltration in liver tissue, including CD3+, CD4+, CD8+, and CD20+ cells (36). This inflammatory response may contribute to liver injury through cytokine-mediated pathways and represents a potential therapeutic target.

Hematologic and oncologic signals

Identifying blast cell proliferation as the strongest signal (ROR = 518.94) requires careful interpretation within the context of the therapeutic indication of dasatinib for hematologic malignancies. This signal may represent disease progression, treatment resistance, or potentially paradoxical effects of dasatinib on blast cell populations. In Philadelphia chromosome–positive leukemia, dasatinib functions primarily through BCR-ABL kinase inhibition; however, its broad kinase inhibitory profile may have unintended consequences on normal hematopoietic cell populations (37, 38).

Primary effusion lymphoma, another highly significant sign, is a rare but serious potential complication. This finding suggests possible interaction between dasatinib and viral-associated lymphoproliferative disorders, particularly given the immunomodulatory effects of dasatinib. This mechanism may involve alterations in immune surveillance or direct effects on lymphocyte populations (39).

The indication for allogeneic bone marrow transplantation therapy likely reflects the clinical context in which dasatinib is used because many patients with advanced hematologic malignancies may require transplantation. However, it also raises questions about the potential interactions between dasatinib and transplant-related complications, including graft-versus-host disease and transplant-related toxicities (40).

Cardiovascular and vascular effects

Our analysis revealed several cardiovascular signals, including peripheral artery stenosis and ejection fraction abnormalities, consistent with emerging concerns regarding tyrosine kinase inhibitor-associated vascular toxicities. Unlike nilotinib and ponatinib, which are more strongly associated with arterial occlusive events, dasatinib appears to have a relatively lower incidence of peripheral arterial disease (7, 41). However, identifying these signals suggests that cardiovascular monitoring remains important, particularly in patients with pre-existing cardiovascular risk factors.

The mechanism of dasatinib-induced cardiovascular effects may involve the off-target inhibition of kinases, which are important for endothelial function and vascular homeostasis. Src family kinases play crucial roles in endothelial cell signaling and vascular integrity, and their inhibition by dasatinib may contribute to endothelial dysfunction (42). Besides, the effects of dasatinib on nitric oxide signaling and inflammatory pathways may contribute to vascular complications (43).

Dermatologic and hypopigmentation effects

Identifying skin hypopigmentation and depigmentation signals provides insight into the effects of dasatinib on melanocyte function and pigment production. This effect is likely due to the inhibition of c-Kit signaling by the drug, which is essential for melanocyte survival and melanin synthesis (44). The mechanism involves the disruption of stem cell factor/c-Kit signaling pathways that regulate melanocyte development, migration, and pigment production.

Despite being generally cosmetic, these dermatological effects can be distressing to patients and may have broader effects on stem cell populations. The reversibility of these effects upon drug discontinuation suggests that the underlying melanocyte population remains intact with functional inhibition, rather than cellular destruction (45).

Novel and unexpected signals

The several unexpected signals identified in our analysis warrant further investigation. Despite being rare, hydrops fetalis raises significant concerns regarding dasatinib use during pregnancy and suggests potential teratogenic effects. The mechanism may involve interference with fetal cardiovascular development or fluid balance regulation, although the rarity of this event limits the current mechanistic understanding (46).

A decrease in drug tolerance is an interesting pharmacodynamic signal that may reflect the development of cellular resistance mechanisms or pharmacokinetic alterations that affect drug absorption or metabolism. This could involve the upregulation of efflux pumps, metabolic enzyme induction, or cellular adaptation mechanisms that reduce drug sensitivity over time (47).

Identifying cytomegalovirus-related complications (CMV enterocolitis and CMV colitis) suggests that the immunosuppressive effects of dasatinib may predispose patients to opportunistic infections. This effect is likely due to the impact of the drug on T cell function and immune surveillance, potentially through Src family kinase inhibition in the immune cells (48).

Subgroup analysis and population-specific considerations

Subgroup analyses revealed important demographic and clinical variations in the adverse event profiles. The consistent appearance of pleural effusion across all demographic strata confirms that it is a universal concern that requires vigilant monitoring, regardless of patient characteristics. However, the enrichment of hematological signals (blast cell proliferation, Philadelphia chromosome-positive) in adult, elderly, and male patients suggests potential age- and sex-related differences in drug sensitivity or disease progression patterns.

The distinctive signals observed in pediatric and low-weight patients, including hemorrhagic enterocolitis and gastrointestinal hemorrhage, highlight the need for age-specific monitoring. These differences may reflect developmental variations in drug metabolism, organ sensitivity, or the underlying disease characteristics. The tendency of the medium- and high-weight groups to report more cardiopulmonary events suggests potential dose–weight relationships that may require therapeutic drug monitoring or dose optimization strategies.

Temporal patterns and clinical monitoring implications

Weibull distribution analysis revealed an early failure pattern (β = 0.64), which may help refine clinical monitoring protocols. The high proportion of early onset events (25.6% within 30 days) supports intensive monitoring during treatment initiation, whereas the significant late-onset component (28.2% after 360 days) indicates the need for sustained vigilance throughout therapy.

This bimodal pattern suggests different underlying mechanisms of early and late AEs. Early events likely represent acute toxicities related to initial drug exposure and rapid pharmacological effects, although contributions from pre-existing comorbidities, more advanced or resistant disease, and baseline organ dysfunction cannot be excluded. Late events may reflect cumulative toxicities, organ dysfunction, or the development of secondary complications. A median time to onset of 131.5 days provides a useful benchmark for clinical monitoring, suggesting that most events occur within the first 4–5 months of therapy, but these time windows should be interpreted as hypothesis-generating given the limited treatment-history data in FAERS.

Comparison with existing clinical practice

Our findings are largely consistent with established clinical knowledge regarding dasatinib safety profiles, while providing measures of disproportionality (signal strength) and novel signal identification. The prominence of pleural effusion aligns with clinical trial data demonstrating incidence rates of 10%–35% in patients treated with dasatinib (49). From a class perspective, pleural and other serosal effusions are more characteristic of dasatinib, whereas arterial occlusive and metabolic events are more prominent with nilotinib and ponatinib, and imatinib generally shows a more favorable cardiometabolic profile, a pattern that has also been observed in other clinical and FAERS-based evaluations of BCR-ABL1 TKIs (10, 19, 20). However, our analysis provides a real-world perspective that may capture AEs not fully represented in controlled clinical trials.

Identifying hepatotoxicity as a major signal is consistent with preclinical and clinical studies demonstrating the potential of dasatinib for liver injury (50, 51). The relative signal strength from our analysis (ROR = 29.17) provide a valuable context for clinical decision-making and patient counseling. Similarly, cardiovascular signals align with emerging concerns regarding tyrosine kinase inhibitor-associated vascular toxicities, although dasatinib appears to have a more favorable cardiovascular profile compared to other agents of this class (52).

The novel signals identified in our analysis, such as alveolar proteinosis and cytomegalovirus-related complications, extend beyond the currently recognized adverse effects and may represent important areas for future clinical investigation. These findings highlight the value of large-scale pharmacovigilance studies in identifying rare but potentially significant AEs that might not be captured in clinical trials.

Limitations and methodological considerations

Several limitations inherent to FAERS-based pharmacovigilance studies must be addressed. The spontaneous reporting nature of FAERS introduces a potential reporting bias, with serious or unexpected events being more likely to be reported than mild or well-known effects. Additionally, the database lacks comprehensive denominator data, preventing the calculation of true incidence rates and limiting risk quantification to disproportionate measures. Information on treatment indication, disease phase, comorbidities, dosing, treatment duration, and prior TKIs is also limited, so residual confounding by disease severity and treatment selection is unavoidable; many hematologic and transplantation-related PTs may therefore reflect the underlying course of advanced leukemia rather than de novo dasatinib toxicity. Despite being addressed through standard deduplication procedures, duplicate reporting may still influence signal strength. Furthermore, the retrospective nature of the analysis and reliance on medical terminology coding may have introduced classification errors or missed relevant events coded using different terms, and all signals should be regarded as hypothesis-generating rather than proof of causality.

Conclusion

This FAERS-based pharmacovigilance analysis confirmed the well-known risks of dasatinib, especially respiratory complications, such as pleural effusion, while identifying important hepatotoxicity, cardiovascular, and novel adverse event signals. The early failure pattern of AEs underscores the need for vigilant monitoring, both during treatment initiation and over the long-term. Subgroup differences suggest that surveillance should be tailored to patient demographics. Mechanistic insights highlight the off-target effects of dasatinib in multiple organ systems. These findings support optimized clinical management and further investigation of rare and emerging toxicities to improve patient safety.

Data availability statement

The original contributions presented in this study are included in this article/Supplementary material, further inquiries can be directed to the corresponding authors.

Ethics statement

Ethical review and approval was not required for the study of human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. This study was a secondary analysis of publicly available, de-identified data from the U.S. FDA Adverse Event Reporting System (FAERS); no identifiable personal information was accessed or used.

Author contributions

ZL: Investigation, Data curation, Software, Writing – original draft, Visualization, Methodology, Conceptualization, Writing – review & editing, Project administration, Validation, Formal analysis, Supervision. GS: Writing – review & editing, Supervision. YZ: Writing – review & editing, Funding acquisition, Supervision. XL: Supervision, Writing – review & editing, Funding acquisition.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was funded by the National Natural Science Foundation of China (82260914, 82260798, and 82460802).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2025.1709089/full#supplementary-material

References

1. Komorowski L, Dabkowska A, Madzio J, Pastorczak A, Szczygiel K, Janowska M, et al. Concomitant inhibition of the thioredoxin system and nonhomologous DNA repair potently sensitizes Philadelphia-positive lymphoid leukemia to tyrosine kinase inhibitors. Hemasphere. (2024) 8:e56. doi: 10.1002/hem3.56

PubMed Abstract | Crossref Full Text | Google Scholar

2. Komorowski L, Fidyt K, Patkowska E, Firczuk M. Philadelphia chromosome-positive leukemia in the lymphoid lineage-similarities and differences with the myeloid lineage and specific vulnerabilities. Int J Mol Sci. (2020) 21:5776. doi: 10.3390/ijms21165776

PubMed Abstract | Crossref Full Text | Google Scholar

3. Hochhaus A, Baccarani M, Silver R, Schiffer C, Apperley J, Cervantes F, et al. European LeukemiaNet 2020 recommendations for treating chronic myeloid leukemia. Leukemia. (2020) 34:966–84. doi: 10.1038/s41375-020-0776-2

PubMed Abstract | Crossref Full Text | Google Scholar

4. Sasaki K, Jabbour E, Short N, Jain N, Ravandi F, Pui C, et al. Acute lymphoblastic leukemia: a population-based study of outcome in the United States based on the surveillance, epidemiology, and end results (SEER) database, 1980-2017. Am J Hematol. (2021) 96:650–8. doi: 10.1002/ajh.26156

PubMed Abstract | Crossref Full Text | Google Scholar

5. Eiring A, Khorashad J, Morley K, Deininger M. Advances in the treatment of chronic myeloid leukemia. BMC Med. (2011) 9:99. doi: 10.1186/1741-7015-9-99

PubMed Abstract | Crossref Full Text | Google Scholar

6. Breccia M, Salaroli A, Molica M, Alimena G. Systematic review of dasatinib in chronic myeloid leukemia. Onco Targets Ther. (2013) 6:257–65. doi: 10.2147/OTT.S35360

PubMed Abstract | Crossref Full Text | Google Scholar

7. Talpaz M, Shah N, Kantarjian H, Donato N, Nicoll J, Paquette R, et al. Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias. N Engl J Med. (2006) 354:2531–41. doi: 10.1056/NEJMoa055229

PubMed Abstract | Crossref Full Text | Google Scholar

8. Lombardo L, Lee F, Chen P, Norris D, Barrish J, Behnia K, et al. Discovery of N-(2-chloro-6-methyl- phenyl)-2-(6-(4-(2-hydroxyethyl)- piperazin-1-yl)-2-methylpyrimidin-4- ylamino)thiazole-5-carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays. J Med Chem. (2004) 47:6658–61. doi: 10.1021/jm049486a

PubMed Abstract | Crossref Full Text | Google Scholar

9. Chang Q, Jorgensen C, Pawson T, Hedley D. Effects of dasatinib on EphA2 receptor tyrosine kinase activity and downstream signalling in pancreatic cancer. Br J Cancer. (2008) 99:1074–82. doi: 10.1038/sj.bjc.6604676

PubMed Abstract | Crossref Full Text | Google Scholar

10. Hughes T, Laneuville P, Rousselot P, Snyder D, Rea D, Shah N, et al. Incidence, outcomes, and risk factors of pleural effusion in patients receiving dasatinib therapy for Philadelphia chromosome-positive leukemia. Haematologica. (2019) 104:93–101. doi: 10.3324/haematol.2018.188987

PubMed Abstract | Crossref Full Text | Google Scholar

11. Montani D, Bergot E, Günther S, Savale L, Bergeron A, Bourdin A, et al. Pulmonary arterial hypertension in patients treated by dasatinib. Circulation. (2012) 125:2128–37. doi: 10.1161/CIRCULATIONAHA.111.079921

PubMed Abstract | Crossref Full Text | Google Scholar

12. Weatherald J, Chaumais M, Savale L, Jaïs X, Seferian A, Canuet M, et al. Long-term outcomes of dasatinib-induced pulmonary arterial hypertension: a population-based study. Eur Respir J. (2017) 50:1700217. doi: 10.1183/13993003.00217-2017

PubMed Abstract | Crossref Full Text | Google Scholar

13. Battini V, Cocco M, Barbieri M, Powell G, Carnovale C, Clementi E, et al. Timing matters: a machine learning method for the prioritization of drug-drug interactions through signal detection in the FDA adverse event reporting system and their relationship with time of co-exposure. Drug Saf. (2024) 47:895–907. doi: 10.1007/s40264-024-01430-8

PubMed Abstract | Crossref Full Text | Google Scholar

14. Kantarjian H, Giles F, Wunderle L, Bhalla K, O’Brien S, Wassmann B, et al. Nilotinib in imatinib-resistant CML and Philadelphia chromosome-positive ALL. N Engl J Med. (2006) 354:2542–51. doi: 10.1056/NEJMoa055104

PubMed Abstract | Crossref Full Text | Google Scholar

15. Mauro M, Deininger M. Management of drug toxicities in chronic myeloid leukaemia. Best Pract Res Clin Haematol. (2009) 22:409–29. doi: 10.1016/j.beha.2009.06.001

PubMed Abstract | Crossref Full Text | Google Scholar

16. Quintás-Cardama A, Kantarjian H, O’Brien S, Borthakur G, Bruzzi J, Munden R, et al. Pleural effusion in patients with chronic myelogenous leukemia treated with dasatinib after imatinib failure. J Clin Oncol. (2007) 25:3908–14. doi: 10.1200/JCO.2007.12.0329

PubMed Abstract | Crossref Full Text | Google Scholar

17. Hughes T, Saglio G, Branford S, Soverini S, Kim D, Müller M, et al. Impact of baseline BCR-ABL mutations on response to nilotinib in patients with chronic myeloid leukemia in chronic phase. J Clin Oncol. (2009) 27:4204–10. doi: 10.1200/JCO.2009.21.8230

PubMed Abstract | Crossref Full Text | Google Scholar

18. Cortes J, Rousselot P, Kim D, Ritchie E, Hamerschlak N, Coutre S, et al. Dasatinib induces complete hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in blast crisis. Blood. (2007) 109:3207–13. doi: 10.1182/blood.V112.11.4290.4290

Crossref Full Text | Google Scholar

19. Moslehi J, Deininger M. Tyrosine kinase inhibitor-associated cardiovascular toxicity in chronic myeloid leukemia. J Clin Oncol. (2015) 33:4210–8. doi: 10.1200/JCO.2015.62.4718

PubMed Abstract | Crossref Full Text | Google Scholar

20. Valent P, Hadzijusufovic E, Schernthaner G, Wolf D, Rea D, le Coutre P. Vascular safety issues in CML patients treated with BCR/ABL1 kinase inhibitors. Blood. (2015) 125:901–6. doi: 10.1182/blood-2014-09-594432

PubMed Abstract | Crossref Full Text | Google Scholar

21. Steegmann J, Baccarani M, Breccia M, Casado L, García-Gutiérrez V, Hochhaus A, et al. European LeukemiaNet recommendations for the management and avoidance of adverse events of treatment in chronic myeloid leukaemia. Leukemia. (2016) 30:1648–71. doi: 10.1038/leu.2016.104

PubMed Abstract | Crossref Full Text | Google Scholar

22. Bergeron A, Godet C, Chevret S, Lorillon G, Peffault de Latour R, de Revel T, et al. Bronchiolitis obliterans syndrome after allogeneic hematopoietic SCT: phenotypes and prognosis. Bone Marrow Transplant. (2013) 48:819–24. doi: 10.1038/bmt.2012.241

PubMed Abstract | Crossref Full Text | Google Scholar

23. Porkka K, Khoury H, Paquette R, Matloub Y, Sinha R, Cortes J. Dasatinib 100 mg once daily minimizes the occurrence of pleural effusion in patients with chronic myeloid leukemia in chronic phase and efficacy is unaffected in patients who develop pleural effusion. Cancer. (2010) 116:377–86. doi: 10.1002/cncr.24734

PubMed Abstract | Crossref Full Text | Google Scholar

24. Mustjoki S, Ekblom M, Arstila T, Dybedal I, Epling-Burnette P, Guilhot F, et al. Clonal expansion of T/NK-cells during tyrosine kinase inhibitor dasatinib therapy. Leukemia. (2009) 23:1398–405. doi: 10.1038/leu.2009.46

PubMed Abstract | Crossref Full Text | Google Scholar

25. Soverini S, Colarossi S, Gnani A, Rosti G, Castagnetti F, Poerio A, et al. Contribution of ABL kinase domain mutations to imatinib resistance in different subsets of Philadelphia-positive patients: by the GIMEMA working party on chronic Myeloid Leukemia. Clin Cancer Res. (2006) 12:7374–9. doi: 10.1158/1078-0432.CCR-06-1516

PubMed Abstract | Crossref Full Text | Google Scholar

26. Shah N, Tran C, Lee F, Chen P, Norris D, Sawyers C. Overriding imatinib resistance with a novel ABL kinase inhibitor. Science. (2004) 305:399–401. doi: 10.1126/science.1099480

PubMed Abstract | Crossref Full Text | Google Scholar

27. Tokarski J, Newitt J, Chang C, Cheng J, Wittekind M, Kiefer S, et al. The structure of Dasatinib (BMS-354825) bound to activated ABL kinase domain elucidates its inhibitory activity against imatinib-resistant ABL mutants. Cancer Res. (2006) 66:5790–7. doi: 10.1158/0008-5472.CAN-05-4187

PubMed Abstract | Crossref Full Text | Google Scholar

28. Bantscheff M, Eberhard D, Abraham Y, Bastuck S, Boesche M, Hobson S, et al. Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors. Nat Biotechnol. (2007) 25:1035–44. doi: 10.1038/nbt1328

PubMed Abstract | Crossref Full Text | Google Scholar

29. Morgan M, Onono F, Spielmann H, Subramanian T, Scherr M, Venturini L, et al. Modulation of anthracycline-induced cytotoxicity by targeting the prenylated proteome in myeloid leukemia cells. J Mol Med. (2012) 90:149–61. doi: 10.1007/s00109-011-0814-7

PubMed Abstract | Crossref Full Text | Google Scholar

30. Rix U, Hantschel O, Dürnberger G, Remsing Rix L, Planyavsky M, Fernbach N, et al. Chemical proteomic profiles of the BCR-ABL inhibitors imatinib, nilotinib, and dasatinib reveal novel kinase and nonkinase targets. Blood. (2007) 110:4055–63. doi: 10.1182/blood-2007-07-102061

PubMed Abstract | Crossref Full Text | Google Scholar

31. Skaggs B, Gorre M, Ryvkin A, Burgess M, Xie Y, Han Y, et al. Phosphorylation of the ATP-binding loop directs oncogenicity of drug-resistant BCR-ABL mutants. Proc Natl Acad Sci U S A. (2006) 103:19466–71. doi: 10.1073/pnas.0609239103

PubMed Abstract | Crossref Full Text | Google Scholar

32. Yang K, Fu LW. Mechanisms of resistance to BCR-ABL TKIs and the therapeutic strategies: a review. Crit Rev Oncol Hematol. (2015) 93:277–92. doi: 10.1016/j.critrevonc.2014.11.001

PubMed Abstract | Crossref Full Text | Google Scholar

33. Wu J, Meng F, Kong L, Peng Z, Ying Y, Bornmann W, et al. Association between imatinib-resistant BCR-ABL mutation-negative leukemia and persistent activation of LYN kinase. J Natl Cancer Inst. (2008) 100:926–39. doi: 10.1093/jnci/djn188

PubMed Abstract | Crossref Full Text | Google Scholar

34. Zhang J, Adrián F, Jahnke W, Cowan-Jacob S, Li A, Iacob R, et al. Targeting Bcr-Abl by combining allosteric with ATP-binding-site inhibitors. Nature. (2010) 463:501–6. doi: 10.1038/nature08675

PubMed Abstract | Crossref Full Text | Google Scholar

35. O’Hare T, Zabriskie M, Eiring A, Deininger M. Pushing the limits of targeted therapy in chronic myeloid leukaemia. Nat Rev Cancer. (2012) 12:513–26. doi: 10.1038/nrc3317

PubMed Abstract | Crossref Full Text | Google Scholar

36. Redaelli S, Piazza R, Rostagno R, Magistroni V, Perini P, Marega M, et al. Activity of bosutinib, dasatinib, and nilotinib against 18 imatinib-resistant BCR/ABL mutants. J Clin Oncol. (2009) 27:469–71. doi: 10.1200/JCO.2008.19.8853

PubMed Abstract | Crossref Full Text | Google Scholar

37. Weisberg E, Manley P, Cowan-Jacob S, Hochhaus A, Griffin J. Second generation inhibitors of BCR-ABL for the treatment of imatinib-resistant chronic myeloid leukaemia. Nat Rev Cancer. (2007) 7:345–56. doi: 10.1038/nrc2126

PubMed Abstract | Crossref Full Text | Google Scholar

38. Gorre M, Mohammed M, Ellwood K, Hsu N, Paquette R, Rao P, et al. Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science. (2001) 293:876–80. doi: 10.1126/science.1062538

PubMed Abstract | Crossref Full Text | Google Scholar

39. Schittenhelm M, Shiraga S, Schroeder A, Corbin A, Griffith D, Lee F, et al. Dasatinib (BMS-354825), a dual SRC/ABL kinase inhibitor, inhibits the kinase activity of wild-type, juxtamembrane, and activation loop mutant KIT isoforms associated with human malignancies. Cancer Res. (2006) 66:473–81. doi: 10.1158/0008-5472.CAN-05-2050

PubMed Abstract | Crossref Full Text | Google Scholar

40. Brave M, Goodman V, Kaminskas E, Farrell A, Timmer W, Pope S, et al. Sprycel for chronic myeloid leukemia and Philadelphia chromosome-positive acute lymphoblastic leukemia resistant to or intolerant of imatinib mesylate. Clin Cancer Res. (2008) 14:352–9. doi: 10.1158/1078-0432.CCR-07-4175

PubMed Abstract | Crossref Full Text | Google Scholar

41. Hochhaus A, Kantarjian H, Baccarani M, Lipton J, Apperley J, Druker B, et al. Dasatinib induces notable hematologic and cytogenetic responses in chronic-phase chronic myeloid leukemia after failure of imatinib therapy. Blood. (2007) 109:2303–9. doi: 10.1182/blood-2006-09-047266

PubMed Abstract | Crossref Full Text | Google Scholar

42. Guilhot F, Apperley J, Kim D, Bullorsky E, Baccarani M, Roboz G, et al. Dasatinib induces significant hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in accelerated phase. Blood. (2007) 109:4143–50. doi: 10.1182/blood-2006-09-046839

PubMed Abstract | Crossref Full Text | Google Scholar

43. Shah N, Kantarjian H, Kim D, Réa D, Dorlhiac-Llacer P, Milone J, et al. Intermittent target inhibition with dasatinib 100 mg once daily preserves efficacy and improves tolerability in imatinib-resistant and -intolerant chronic-phase chronic myeloid leukemia. J Clin Oncol. (2008) 26:3204–12. doi: 10.1200/JCO.2007.14.9260

PubMed Abstract | Crossref Full Text | Google Scholar

44. Kantarjian H, Shah N, Cortes J, Baccarani M, Agarwal M, Undurraga M, et al. Dasatinib or imatinib in newly diagnosed chronic-phase chronic myeloid leukemia: 2-year follow-up from a randomized phase 3 trial (DASISION). Blood. (2012) 119:1123–9. doi: 10.1182/blood-2011-08-376087

PubMed Abstract | Crossref Full Text | Google Scholar

45. Cross N, White H, Müller M, Saglio G, Hochhaus A. Standardized definitions of molecular response in chronic myeloid leukemia. Leukemia. (2012) 26:2172–5. doi: 10.1038/leu.2012.104

PubMed Abstract | Crossref Full Text | Google Scholar

46. Hughes T, Hochhaus A, Branford S, Müller M, Kaeda J, Foroni L, et al. Long-term prognostic significance of early molecular response to imatinib in newly diagnosed chronic myeloid leukemia: an analysis from the International Randomized Study of Interferon and STI571 (IRIS). Blood. (2010) 116:3758–65. doi: 10.1182/blood-2010-03-273979

PubMed Abstract | Crossref Full Text | Google Scholar

47. Baccarani M, Deininger M, Rosti G, Hochhaus A, Soverini S, Apperley J, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. (2013) 122:872–84. doi: 10.1182/blood-2013-05-501569

PubMed Abstract | Crossref Full Text | Google Scholar

48. Saglio G, Kim D, Issaragrisil S, le Coutre P, Etienne G, Lobo C, et al. Nilotinib versus imatinib for newly diagnosed chronic myeloid leukemia. N Engl J Med. (2010) 362:2251–9. doi: 10.1056/NEJMoa0912614

PubMed Abstract | Crossref Full Text | Google Scholar

49. Mahon F, Réa D, Guilhot J, Guilhot F, Huguet F, Nicolini F, et al. Discontinuation of imatinib in patients with chronic myeloid leukaemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol. (2010) 11:1029–35. doi: 10.1016/S1470-2045(10)70233-3

PubMed Abstract | Crossref Full Text | Google Scholar

50. Ross D, Branford S, Seymour J, Schwarer A, Arthur C, Yeung D, et al. Safety and efficacy of imatinib cessation for CML patients with stable undetectable minimal residual disease: results from the TWISTER study. Blood. (2013) 122:515–22. doi: 10.1182/blood-2013-02-483750

PubMed Abstract | Crossref Full Text | Google Scholar

51. Etienne G, Guilhot J, Rea D, Rigal-Huguet F, Nicolini F, Charbonnier A, et al. Long-Term follow-up of the french Stop imatinib (STIM1) study in patients with chronic myeloid leukemia. J Clin Oncol. (2017) 35:298–305. doi: 10.1200/JCO.2016.68.2914

PubMed Abstract | Crossref Full Text | Google Scholar

52. Saussele S, Richter J, Guilhot J, Gruber F, Hjorth-Hansen H, Almeida A, et al. Discontinuation of tyrosine kinase inhibitor therapy in chronic myeloid leukaemia (EURO-SKI): a prespecified interim analysis of a prospective, multicentre, non-randomised, trial. Lancet Oncol. (2018) 19:747–57. doi: 10.1016/S1470-2045(18)30192-X

PubMed Abstract | Crossref Full Text | Google Scholar