Introduction
Drug-induced myelosuppression (DIM) is a common and serious adverse reaction in drug therapy, involving the suppression of hematopoietic function in the blood system, leading to reductions in white blood cells, red blood cells, and platelets. In severe cases, it can result in complications such as infections, anemia, and bleeding. 1 Myelosuppression is a known side effect of many drugs, particularly chemotherapeutic agents, immunosuppressants, and certain targeted therapies. The clinical manifestations of DIM are often subtle, typically emerging in the later stages of treatment, posing significant challenges to patients’ health and treatment outcomes. 2 With the increasing variety of medications and the continuous development of therapeutic methods, DIM has garnered growing attention in both clinical practice and drug regulation.
Statistically, DIM is especially common in cancer treatment, with several chemotherapeutic agents such as cyclophosphamide and paclitaxel shown to induce myelosuppression by directly affecting hematopoietic cells in the bone marrow. 3 , 4 Additionally, some targeted therapies and immunotherapies, including certain BCL-2 inhibitors like Venetoclax and immune checkpoint inhibitors, have also been associated with myelosuppression-related adverse reactions in clinical applications. 5 , 6 While these drugs offer significant therapeutic benefits for major diseases like cancer, their adverse effects can impact patients’ treatment plans and quality of life, potentially leading to treatment interruption or even death.
Risk assessment and early identification of drug-induced myelosuppression are crucial for clinical treatment. However, due to the diversity and complexity of adverse reactions, many potential DIM risks remain underrecognized, especially during long-term post-market drug use. Therefore, post-market safety monitoring and pharmacovigilance have become critical tools for identifying and managing these risks. The FDA Adverse Event Reporting System (FAERS) provides a valuable database that records global drug adverse event reports, including those related to DIM. 7 This study leverages the FAERS database to conduct signal mining and thoroughly analyze the potential risks of drug-induced myelosuppression. We will apply common signal detection algorithms, such as the Disproportionality Analysis (DPA) and Bayesian models, to identify drug signals associated with myelosuppression and further explore its occurrence patterns and risk factors. Through this research, we aim to provide scientific evidence on drug safety for regulatory agencies, clinicians, and patients, enhancing drug safety monitoring and promoting the optimization of personalized treatment regimens to reduce the risk of DIM.
Materials and Methods
Data Source and Process
Adverse event data for this study were obtained from the FDA Adverse Event Reporting System (FAERS) database ( https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html ). This database has been publicly available since 2004, and the data for this study were downloaded from the FAERS database covering adverse event reports from the first quarter of 2004 to the third quarter of 2024. This study used publicly available data from the FDA Adverse Event Reporting System (FAERS) and did not involve human participants, human tissue, or identifiable private information. Therefore, ethical approval and informed consent were not required, in accordance with institutional and international guidelines.
For data with the same “caseid” (report code) from the FAERS database, only the most recent report based on the date was retained, and duplicate reports were removed. Drug names were standardized according to the RxNorm drug normalization nomenclature to ensure consistency across the FAERS data. The International Medical Terminology Dictionary, version 27.1 (MedDRA 27.1), was used to match the primary term (PT) for “MYELOSUPPRESSION” adverse events.
After standardizing drug and adverse event names, reports of myelosuppression and the primary suspected (PS) drugs were collected. These reports were characterized by gender, age, weight, indication, reporting country, and outcome. This study complies with the RECORD guidelines. 8
Signal Analysis Algorithms
In this study, disproportionality analysis (DPA), a commonly used algorithm in pharmacovigilance research, was employed to detect potential signals of drug-related myelosuppression adverse events. The disproportionality analysis is a widely used data mining technique that compares the observed frequency ratios between exposed and unexposed populations using a 2 × 2 contingency table to analyze the association between drugs and adverse events ( Table 1 ). Table 1. Two-By-Two Contingency Table for Disproportionality Analysis of DIM-Related Adverse Events. DIM-Related adverse events Other adverse events Total Suspected drug a b a + b Other drugs c d c + d Total a + c b + d a + b + c+d Abbreviation: a, number of reports containing both the target drug and target adverse drug reaction; b, number of reports containing other adverse drug reaction of the target drug; c, number of reports containing the target adverse drug reaction of other drugs; d, number of reports containing other drugs and other adverse drug reactions.
Table 1.
Two-By-Two Contingency Table for Disproportionality Analysis of DIM-Related Adverse Events.
Abbreviation: a, number of reports containing both the target drug and target adverse drug reaction; b, number of reports containing other adverse drug reaction of the target drug; c, number of reports containing the target adverse drug reaction of other drugs; d, number of reports containing other drugs and other adverse drug reactions.
| DIM-Related adverse events | Other adverse events | Total | |
| Suspected drug | a | b | a + b |
| Other drugs | c | d | c + d |
| Total | a + c | b + d | a + b + c+d |
For this study, the following signal detection algorithms were used to calculate signal strength: Reporting Odds Ratio (ROR), 9 Proportional Reporting Ratio (PRR), 10 Bayesian Confidence Propagation Neural Network (BCPNN), 11 and Empirical Bayesian Geometric Mean (EBGM). 12 Adverse event signals were considered valid if they met the positive signal criteria for these four algorithms ( Table 2 ). After examining adverse events not mentioned in the FDA drug labeling ( https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm ), valid signals were classified as new adverse event signals. Table 2. Four Major Algorithms Used for Signal Detection. Algorithms Equation Criteria ROR ROR = ad/b/c Lower limit of 95% CI > 1, N ≥3 95%CI = e ln(ROR)±1.96(1/a+1/b+1/c+1/d)^0.5 PRR PRR = a(c + d)/c/(a + b) PRR ≥2, χ 2 ≥ 4, N ≥3 χ 2 = [(ad-bc)^2](a + b + c+d)/[(a + b)(c + d)(a + c)(b + d)] BCPNN IC = log 2 a(a + b + c+d)(a + c)(a + b) IC025>0 95%CI = E(IC) ± 2V(IC)^0.5 EBGM EBGM = a(a + b + c+d)/(a + c)/(a + b) EBGM05>2 95%CI = e ln(EBGM)±1.96(1/a+1/b+1/c+1/d)^0.5 Abbreviation: a, number of reports containing both the target drug and target adverse drug reaction; b, number of reports containing other adverse drug reaction of the target drug; c, number of reports containing the target adverse drug reaction of other drugs; d, number of reports containing other drugs and other adverse drug reactions. 95%CI, 95% confidence interval; N, the number of reports; χ2, chi-squared; IC, information component; IC025, the lower limit of 95% CI of the IC; E(IC), the IC expectations; V(IC), the variance of IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower limit of 95% CI of EBGM.
Table 2.
Four Major Algorithms Used for Signal Detection.
Abbreviation: a, number of reports containing both the target drug and target adverse drug reaction; b, number of reports containing other adverse drug reaction of the target drug; c, number of reports containing the target adverse drug reaction of other drugs; d, number of reports containing other drugs and other adverse drug reactions. 95%CI, 95% confidence interval; N, the number of reports; χ2, chi-squared; IC, information component; IC025, the lower limit of 95% CI of the IC; E(IC), the IC expectations; V(IC), the variance of IC; EBGM, empirical Bayesian geometric mean; EBGM05, the lower limit of 95% CI of EBGM.
| Algorithms | Equation | Criteria |
| ROR | ROR = ad/b/c | Lower limit of 95% CI > 1, N ≥3 |
| 95%CI = e ln(ROR)±1.96(1/a+1/b+1/c+1/d)^0.5 | ||
| PRR | PRR = a(c + d)/c/(a + b) | PRR ≥2, χ 2 ≥ 4, N ≥3 |
| χ 2 = [(ad-bc)^2](a + b + c+d)/[(a + b)(c + d)(a + c)(b + d)] | ||
| BCPNN | IC = log 2 a(a + b + c+d)(a + c)(a + b) | IC025>0 |
| 95%CI = E(IC) ± 2V(IC)^0.5 | ||
| EBGM | EBGM = a(a + b + c+d)/(a + c)/(a + b) | EBGM05>2 |
| 95%CI = e ln(EBGM)±1.96(1/a+1/b+1/c+1/d)^0.5 |
All data in this study were processed and statistically analyzed using R version 4.4.0 and MS Excel. The data extraction and analysis workflow is shown in Figure 1 . Figure 1. Analysis Process of Drug-Induced Bone Marrow Suppression Signals.
Results
Basic Characteristics of Adverse Events Related to Myelosuppression
As of the third quarter of 2024, the FAERS database contained a total of 21 380 adverse event reports related to myelosuppression. The number of reports related to myelosuppression was relatively low before 2010, but has shown a continuous and accelerating upward trend since then, particularly after 2015. In 2021, the number of reports peaked at 3501 cases ( Figure 2 ). To further explore the trend, a polynomial fitting curve was plotted, showing a rapid upward trajectory. The coefficient of determination (R 2 = 0.938) indicates that the model explains 93.80% of the data variability, suggesting that the trend has high reference value. This sustained increase may be attributed to multiple factors, including the expanded clinical application of new myelosuppressive agents (e.g., targeted therapies and immunotherapies), increased global use of cancer treatments, and the gradual improvement of pharmacovigilance systems, especially in countries such as China where spontaneous reporting has become more standardized in recent years. Figure 2. Reporting Trends of Adverse Events in Drug Induced Bone Marrow Inhibition.
Table 3 presents the demographic characteristics of the population associated with myelosuppression adverse events. Of the reports, 50.2% were from female patients, 35.5% from male patients, and 14.3% had an unknown gender. The age distribution revealed that the largest proportion of adverse events occurred in patients aged 18-65 years (44.42%) among those with known age information. The highest proportion of myelosuppression adverse events was observed in patients weighing 50-70 kg (26.07%), although 56.33% of the reports did not provide weight information. Notably, the highest incidence of myelosuppression was observed in patients with breast cancer (10.6%), followed by acute myeloid leukemia (5.2%) and cases with an unknown indication (3.7%). Geographically, the highest number of reports were from China (57.40%), followed by Japan (12.40%) and the United States (6.76%). Table 3. Baseline Characteristics of Drug-Induced Bone Marrow Suppression Population. Characteristics Case numbers Case proportion (%) Number of events 21 380 - Gender Female 10 725 50.2% Male 7594 35.5% Miss 3061 14.3% Age Median (IQR) 57 <18 1511 7.07% 18-65 9497 44.42% 65-85 4954 23.17% >85 215 1.01% Miss 5203 24.34% Weight(KG) <50 2077 9.71% 50-70 5574 26.07% 70-90 1447 6.77% ≥90 238 1.11% Miss 12 044 56.33% Top 5 indication Breast cancer 2271 10.6% Acute myeloid leukaemia 1102 5.2% Product used for unknown indication 781 3.7% Medication dilution 728 3.4% Plasma cell myeloma 670 3.1% Top 5 Reported countries China 12 272 57.40% Japan 2652 12.40% United STATES 1445 6.76% Germany 750 3.51% Canada 444 2.08% Abbreviation: interquartile range, IQR.
Table 3.
Baseline Characteristics of Drug-Induced Bone Marrow Suppression Population.
Abbreviation: interquartile range, IQR.
| Characteristics | Case numbers | Case proportion (%) |
| Number of events | 21 380 | - |
| Gender | ||
| Female | 10 725 | 50.2% |
| Male | 7594 | 35.5% |
| Miss | 3061 | 14.3% |
| Age | ||
| Median (IQR) | 57 | |
| <18 | 1511 | 7.07% |
| 18-65 | 9497 | 44.42% |
| 65-85 | 4954 | 23.17% |
| >85 | 215 | 1.01% |
| Miss | 5203 | 24.34% |
| Weight(KG) | ||
| <50 | 2077 | 9.71% |
| 50-70 | 5574 | 26.07% |
| 70-90 | 1447 | 6.77% |
| ≥90 | 238 | 1.11% |
| Miss | 12 044 | 56.33% |
| Top 5 indication | ||
| Breast cancer | 2271 | 10.6% |
| Acute myeloid leukaemia | 1102 | 5.2% |
| Product used for unknown indication | 781 | 3.7% |
| Medication dilution | 728 | 3.4% |
| Plasma cell myeloma | 670 | 3.1% |
| Top 5 Reported countries | ||
| China | 12 272 | 57.40% |
| Japan | 2652 | 12.40% |
| United STATES | 1445 | 6.76% |
| Germany | 750 | 3.51% |
| Canada | 444 | 2.08% |
Drug Analysis
The top 30 drugs associated with the highest number of drug-induced myelosuppression adverse events are shown in Figure 3 . The top five drugs were TRASTUZUMAB (1130 reports), BEVACIZUMAB (1062 reports), VENETOCLAX (982 reports), METHOTREXATE (803 reports), and PERTUZUMAB (709 reports). Among these, BEVACIZUMAB and PERTUZUMAB were noted as not explicitly mentioning the risk of drug-induced myelosuppression in their prescribing information. Of the top 30 drugs with the highest number of myelosuppression-related adverse events, five drugs did not include the risk of myelosuppression in their labels. Figure 3. The Top 30 Drugs Reported for Drug-Induced Bone Marrow Suppression. Note: * Indicates New Signals Not Mentioned in the Manual.
Signal Detection of Drug-Induced Myelosuppression Adverse Events
The statistical results of disproportionality analysis revealed that among the top 30 drugs with the highest signal strength for adverse events, 12 drugs had not mentioned the risk of drug-induced myelosuppression in their labels, indicating a new adverse event signal. The top five drugs with the highest signal strength were TISLELIZUMAB [n = 596, ROR = 694.31 (623.06-773.72)], TORIPALIMAB [n = 192, ROR = 664.4 (550.13-802.41)], GLUCOSE* [n = 48, ROR = 215.08 (156.69-295.23)], CERALASERTIB [n = 1, ROR = 170.79 (19.95-1462)], and DEXKETOPROFEN* [n = 1, ROR = 106.75 (13.35-853.53)].
The five drugs with the highest number of reports, which did not mention the adverse event in their labels, were PERTUZUMAB* [n = 709, ROR = 96.41 (89.11-104.3)], SODIUM CHLORIDE* [n = 668, ROR = 87.73 (80.93-95.1)], MESNA* [n = 56, ROR = 77.33 (58.81-101.68)], FRUQUINTINIB* [n = 49, ROR = 49.98 (37.46-66.69)], and GLUCOSE* [n = 48, ROR = 215.08 (156.69-295.23)].
A total of 10 drugs, including TRASTUZUMAB, PERTUZUMAB, SODIUM CHLORIDE, CYTARABINE, TISLELIZUMAB, ERIBULIN, OBINUTUZUMAB, GILTERITINIB, EPIRUBICIN, and TORIPALIMAB, reported more than 100 myelosuppression-related adverse events ( Table 4 ). Notably, PERTUZUMAB* and SODIUM CHLORIDE* were not mentioned in the prescribing information as being associated with myelosuppression-related adverse events. Table 4. Top 30 Drugs in Signal Strength Ranking. Drug name Case reports ROR (95%CI) PRR (95%CI) EBGM(EBGM05) IC(IC025) TISLELIZUMAB 596 694.31 (623.06-773.72) 388.23 (224037.9) 377.43 (344.74) 8.56 (8.42) TORIPALIMAB 192 664.4 (550.13-802.41) 375.57 (71166.02) 372.21 (317.84) 8.54 (8.29) GLUCOSE* 48 215.08 (156.69-295.23) 172.08 (8155.52) 171.7 (131.73) 7.42 (6.97) CERALASERTIB 1 170.79 (19.95-1462) 142.49 (140.66) 142.49 (23.63) 7.15 (4.84) DEXKETOPROFEN* 1 106.75 (13.35-853.53) 95 (93.11) 94.99 (16.68) 6.57 (4.33) TRILACICLIB* 18 101.88 (62.48-166.12) 91.13 (1605.11) 91.06 (60.48) 6.51 (5.81) PERTUZUMAB* 709 96.41 (89.11-104.3) 87.02 (58353.03) 84.16 (78.8) 6.4 (6.28) GILTERITINIB 247 93.99 (82.37-107.25) 84.86 (20257.7) 83.9 (75.12) 6.39 (6.2) DEXRAZOXANE 33 89.6 (62.58-128.27) 81.19 (2612.83) 81.07 (60.04) 6.34 (5.82) SODIUM CHLORIDE* 668 87.73 (80.93-95.1) 79.88 (50465.02) 77.41 (72.36) 6.27 (6.16) ERIBULIN 349 86.35 (77.29-96.47) 78.62 (26340.65) 77.36 (70.51) 6.27 (6.11) MESNA* 56 77.33 (58.81-101.68) 71.01 (3859.48) 70.82 (56.32) 6.15 (5.75) AMPIROXICAM* 1 56.93 (7.52-431.02) 53.44 (51.51) 53.43 (9.82) 5.74 (3.58) BEPRIDIL 1 53.37 (7.08-402.49) 50.29 (48.37) 50.29 (9.27) 5.65 (3.5) CYTARABINE 654 52.05 (48.05-56.39) 49.2 (29973.27) 47.73 (44.63) 5.58 (5.46) FRUQUINTINIB* 49 49.98 (37.46-66.69) 47.28 (2217.3) 47.17 (37.06) 5.56 (5.14) ZANUBRUTINIB 66 40.38 (31.53-51.7) 38.61 (2413.01) 38.49 (31.3) 5.27 (4.91) OBINUTUZUMAB 287 38.84 (34.48-43.75) 37.21 (9989.35) 36.73 (33.24) 5.2 (5.02) PIPOBROMAN* 2 32.85 (8-134.86) 31.67 (59.46) 31.66 (9.71) 4.98 (3.27) TRASTUZUMAB 1130 32.04 (30.15-34.05) 30.97 (31079.76) 29.39 (27.93) 4.88 (4.79) LORACARBEF* 1 30.5 (4.15-224.17) 29.48 (27.55) 29.48 (5.55) 4.88 (2.77) ENFORTUMAB VEDOTIN 93 29.03 (23.6-35.71) 28.11 (2424.13) 28 (23.54) 4.81 (4.5) POLATUZUMAB VEDOTIN 71 28.89 (22.8-36.62) 27.98 (1843.47) 27.89 (22.88) 4.8 (4.46) EPIRUBICIN 207 28.63 (24.91-32.9) 27.74 (5289.87) 27.48 (24.46) 4.78 (4.58) ALANINE* 1 28.47 (3.88-208.75) 27.58 (25.64) 27.58 (5.21) 4.79 (2.68) IOBENGUANE (131 I) 2 26.69 (6.53-109.04) 25.91 (47.95) 25.91 (7.98) 4.7 (2.99) IDARUBICIN 34 26.6 (18.91-37.44) 25.83 (811.24) 25.79 (19.38) 4.69 (4.19) IFOSFAMIDE 85 25.06 (20.19-31.11) 24.38 (1900.48) 24.29 (20.27) 4.6 (4.29) CHLOROTHIAZIDE* 1 24.4 (3.34-178.11) 23.75 (21.82) 23.75 (4.5) 4.57 (2.47) DECITABINE 69 23.69 (18.64-30.1) 23.08 (1454.38) 23.01 (18.83) 4.52 (4.17) Note: * indicates new signals not mentioned in the manual.
Table 4.
Top 30 Drugs in Signal Strength Ranking.
Note: * indicates new signals not mentioned in the manual.
| Drug name | Case reports | ROR (95%CI) | PRR (95%CI) | EBGM(EBGM05) | IC(IC025) |
| TISLELIZUMAB | 596 | 694.31 (623.06-773.72) | 388.23 (224037.9) | 377.43 (344.74) | 8.56 (8.42) |
| TORIPALIMAB | 192 | 664.4 (550.13-802.41) | 375.57 (71166.02) | 372.21 (317.84) | 8.54 (8.29) |
| GLUCOSE* | 48 | 215.08 (156.69-295.23) | 172.08 (8155.52) | 171.7 (131.73) | 7.42 (6.97) |
| CERALASERTIB | 1 | 170.79 (19.95-1462) | 142.49 (140.66) | 142.49 (23.63) | 7.15 (4.84) |
| DEXKETOPROFEN* | 1 | 106.75 (13.35-853.53) | 95 (93.11) | 94.99 (16.68) | 6.57 (4.33) |
| TRILACICLIB* | 18 | 101.88 (62.48-166.12) | 91.13 (1605.11) | 91.06 (60.48) | 6.51 (5.81) |
| PERTUZUMAB* | 709 | 96.41 (89.11-104.3) | 87.02 (58353.03) | 84.16 (78.8) | 6.4 (6.28) |
| GILTERITINIB | 247 | 93.99 (82.37-107.25) | 84.86 (20257.7) | 83.9 (75.12) | 6.39 (6.2) |
| DEXRAZOXANE | 33 | 89.6 (62.58-128.27) | 81.19 (2612.83) | 81.07 (60.04) | 6.34 (5.82) |
| SODIUM CHLORIDE* | 668 | 87.73 (80.93-95.1) | 79.88 (50465.02) | 77.41 (72.36) | 6.27 (6.16) |
| ERIBULIN | 349 | 86.35 (77.29-96.47) | 78.62 (26340.65) | 77.36 (70.51) | 6.27 (6.11) |
| MESNA* | 56 | 77.33 (58.81-101.68) | 71.01 (3859.48) | 70.82 (56.32) | 6.15 (5.75) |
| AMPIROXICAM* | 1 | 56.93 (7.52-431.02) | 53.44 (51.51) | 53.43 (9.82) | 5.74 (3.58) |
| BEPRIDIL | 1 | 53.37 (7.08-402.49) | 50.29 (48.37) | 50.29 (9.27) | 5.65 (3.5) |
| CYTARABINE | 654 | 52.05 (48.05-56.39) | 49.2 (29973.27) | 47.73 (44.63) | 5.58 (5.46) |
| FRUQUINTINIB* | 49 | 49.98 (37.46-66.69) | 47.28 (2217.3) | 47.17 (37.06) | 5.56 (5.14) |
| ZANUBRUTINIB | 66 | 40.38 (31.53-51.7) | 38.61 (2413.01) | 38.49 (31.3) | 5.27 (4.91) |
| OBINUTUZUMAB | 287 | 38.84 (34.48-43.75) | 37.21 (9989.35) | 36.73 (33.24) | 5.2 (5.02) |
| PIPOBROMAN* | 2 | 32.85 (8-134.86) | 31.67 (59.46) | 31.66 (9.71) | 4.98 (3.27) |
| TRASTUZUMAB | 1130 | 32.04 (30.15-34.05) | 30.97 (31079.76) | 29.39 (27.93) | 4.88 (4.79) |
| LORACARBEF* | 1 | 30.5 (4.15-224.17) | 29.48 (27.55) | 29.48 (5.55) | 4.88 (2.77) |
| ENFORTUMAB VEDOTIN | 93 | 29.03 (23.6-35.71) | 28.11 (2424.13) | 28 (23.54) | 4.81 (4.5) |
| POLATUZUMAB VEDOTIN | 71 | 28.89 (22.8-36.62) | 27.98 (1843.47) | 27.89 (22.88) | 4.8 (4.46) |
| EPIRUBICIN | 207 | 28.63 (24.91-32.9) | 27.74 (5289.87) | 27.48 (24.46) | 4.78 (4.58) |
| ALANINE* | 1 | 28.47 (3.88-208.75) | 27.58 (25.64) | 27.58 (5.21) | 4.79 (2.68) |
| IOBENGUANE (131 I) | 2 | 26.69 (6.53-109.04) | 25.91 (47.95) | 25.91 (7.98) | 4.7 (2.99) |
| IDARUBICIN | 34 | 26.6 (18.91-37.44) | 25.83 (811.24) | 25.79 (19.38) | 4.69 (4.19) |
| IFOSFAMIDE | 85 | 25.06 (20.19-31.11) | 24.38 (1900.48) | 24.29 (20.27) | 4.6 (4.29) |
| CHLOROTHIAZIDE* | 1 | 24.4 (3.34-178.11) | 23.75 (21.82) | 23.75 (4.5) | 4.57 (2.47) |
| DECITABINE | 69 | 23.69 (18.64-30.1) | 23.08 (1454.38) | 23.01 (18.83) | 4.52 (4.17) |
Figure 4 displays a Venn diagram of four algorithms—ROR, PRR, MGPS, and BCPNN—where 129 drugs met the positive signal criteria for all four algorithms. To further investigate more rigorous adverse event signals, we plotted a forest plot ( Figure 5 ) for the drugs that met the positive criteria for all four algorithms and had the top 30 signal strengths. Among these, TRASTUZUMAB, VENETOCLAX, PERTUZUMAB*, CYCLOPHOSPHAMIDE, CYTARABINE, TISLELIZUMAB, ERIBULIN, DARATUMUMAB, OBINUTUZUMAB, GILTERITINIB, ABEMACICLIB, EPIRUBICIN, TORIPALIMAB, BRENTUXIMAB VEDOTIN, and TRASTUZUMAB EMTANSINE each had more than 100 myelosuppression-related adverse events. GLUCOSE*, TRILACICLIB*, PERTUZUMAB*, MESNA*, FRUQUINTINIB*, NIZATIDINE*, and NECITUMUMAB* were identified as having no explicit mention of drug-induced myelosuppression in their prescribing information. Figure 4. Venn Diagrams of Drugs Under Four Algorithms: ROR, PRR, MGPS, and BCPNN. Figure 5. Forest Map of Positive Drugs With the Top 20 Signal Strengths Under the ROR Algorithm. Note: * Indicates New Signals Not Mentioned in the Manual.
Induction Time Analysis
The analysis of the induction time for adverse drug reactions is crucial for drug safety monitoring, clinical medication guidance, regulatory decision-making, and drug development improvement. In this study, the median induction time for adverse events related to PALBOCICLIB was the shortest at 180 days. The Weibull distribution shape parameter (β) was found to be < 1, with a 95% confidence interval (CI) also <1, indicating that the adverse event rate is considered to decrease over time (early failure-type curve). Among the drugs analyzed, TRASTUZUMAB, BEVACIZUMAB*, VENETOCLAX, METHOTREXATE, and PERTUZUMAB* were identified as having myelosuppression induction times following the early failure-type curve, as shown in Table 5 . Table 5. Analysis of the Occurrence Time and Weibull Distribution of Drug-Induced Bone Marrow Suppression Adverse Events. PT Date of onset (days) Weibull distribution Case reports Median(d)(IQR) Scale parameter: α(95%CI) Shape parameter: β(95%CI) Type TRASTUZUMAB 1130 1038 14.44 (13.08∼15.80) 0.73 (0.70∼0.75) Early failure BEVACIZUMAB* 1062 630 17.80 (15.54∼20.05) 0.72 (0.68∼0.76) Early failure VENETOCLAX 982 3064.5 33.53 (28.61∼38.44) 0.63 (0.60∼0.66) Early failure METHOTREXATE 830 4352.5 45.10 (30.81∼59.39) 0.41 (0.38∼0.45) Early failure PERTUZUMAB* 709 180 12.89 (11.79∼14.00) 0.99 (0.94∼1.04) Early failure CYCLOPHOSPHAMIDE 708 1278.5 13.05 (11.40∼14.71) 0.71 (0.67∼0.74) Early failure RITUXIMAB 668 3657.5 12.20 (10.89∼13.52) 0.73 (0.70∼0.76) Early failure CYTARABINE 660 467.5 28.88 (23.08∼34.68) 0.63 (0.58∼0.67) Early failure TISLELIZUMAB 654 155.5 12.29 (11.07∼13.51) 0.90 (0.85∼0.94) Early failure PEMBROLIZUMAB 596 326.5 15.11 (13.53∼16.70) 0.83 (0.79∼0.88) Early failure Note: * indicates new signals not mentioned in the manual; CI, Confidence interval; IQR, Interquartile range.
Table 5.
Analysis of the Occurrence Time and Weibull Distribution of Drug-Induced Bone Marrow Suppression Adverse Events.
Note: * indicates new signals not mentioned in the manual; CI, Confidence interval; IQR, Interquartile range.
| PT | Date of onset (days) | Weibull distribution | |||
| Case reports | Median(d)(IQR) | Scale parameter: α(95%CI) | Shape parameter: β(95%CI) | Type | |
| TRASTUZUMAB | 1130 | 1038 | 14.44 (13.08∼15.80) | 0.73 (0.70∼0.75) | Early failure |
| BEVACIZUMAB* | 1062 | 630 | 17.80 (15.54∼20.05) | 0.72 (0.68∼0.76) | Early failure |
| VENETOCLAX | 982 | 3064.5 | 33.53 (28.61∼38.44) | 0.63 (0.60∼0.66) | Early failure |
| METHOTREXATE | 830 | 4352.5 | 45.10 (30.81∼59.39) | 0.41 (0.38∼0.45) | Early failure |
| PERTUZUMAB* | 709 | 180 | 12.89 (11.79∼14.00) | 0.99 (0.94∼1.04) | Early failure |
| CYCLOPHOSPHAMIDE | 708 | 1278.5 | 13.05 (11.40∼14.71) | 0.71 (0.67∼0.74) | Early failure |
| RITUXIMAB | 668 | 3657.5 | 12.20 (10.89∼13.52) | 0.73 (0.70∼0.76) | Early failure |
| CYTARABINE | 660 | 467.5 | 28.88 (23.08∼34.68) | 0.63 (0.58∼0.67) | Early failure |
| TISLELIZUMAB | 654 | 155.5 | 12.29 (11.07∼13.51) | 0.90 (0.85∼0.94) | Early failure |
| PEMBROLIZUMAB | 596 | 326.5 | 15.11 (13.53∼16.70) | 0.83 (0.79∼0.88) | Early failure |
Discussion
The occurrence of drug-induced myelosuppression involves multiple factors, primarily including the type of drug, its mechanism of action, as well as the individual characteristics and health status of the patient. 13 In recent years, with the increasing frequency of drug usage, drug-induced myelosuppression has gradually attracted widespread clinical attention. Chemotherapy drugs, targeted therapies, and other medications can directly or indirectly affect hematopoietic function in the bone marrow, inhibiting the proliferation and differentiation of hematopoietic stem cells, leading to reductions in white blood cells, red blood cells, and platelets, thereby triggering myelosuppression. Chemotherapy agents such as cyclophosphamide, cytarabine, and anthracyclines (e.g., doxorubicin) have potent cytotoxic effects, directly acting on hematopoietic cells in the bone marrow to suppress their normal function, resulting in extensive myelosuppression. 14 , 15 Additionally, targeted therapies like BCL-2 inhibitors (e.g., venetoclax) and FLT3 inhibitors (e.g., gilteritinib) indirectly impact hematopoiesis in the bone marrow by inhibiting the growth and survival of cancer cells, leading to adverse effects. 16 Immunotherapy agents such as PD-1 inhibitors (e.g., toripalimab) and anti-CD20 monoclonal antibodies (e.g., obinutuzumab) activate the immune system and may induce immune-mediated myelosuppression. 17
Susceptible Populations and Clinical Management
According to data from the FAERS database up to the third quarter of 2024, a total of 21 380 adverse event reports related to myelosuppression were recorded. Notably, since 2019, the number of such reports has significantly increased, reaching 3501 reports in 2021, indicating a rapid upward trend in the occurrence of myelosuppression. The polynomial fitting model (R 2 = 0.938) indicates a strong upward trend in DIM-related reports over time. However, this trend may not solely reflect an increase in true drug-induced myelosuppression risk. It is also likely influenced by improved pharmacovigilance infrastructure, increased clinician awareness, and heightened reporting activity, especially in recent years with the expansion of post-marketing surveillance systems.
In this study, women and patients aged 18-65 years showed a higher frequency of reported drug-induced myelosuppression. Although causality cannot be established from FAERS data, these findings suggest that additional clinical attention may be needed for these subgroups. Clinicians may consider closer monitoring, including more frequent blood tests and early evaluation of hematologic parameters. Pre-treatment risk assessments could also help identify susceptible individuals. While dose adjustment based on age or sex requires further prospective validation, our results highlight the need for individualized pharmacovigilance strategies to enhance treatment safety in high-risk populations.
Additionally, the influence of weight on myelosuppression adverse events was significant, with the highest proportion of patients falling within the 50-70 kg weight range (26.07%). However, 56.33% of patients did not provide weight information. This data suggests that although heavier patients may face a higher risk of adverse drug reactions, missing weight data remains a potential issue. Regarding the distribution of adverse events by indication, breast cancer (10.6%) had the highest incidence of myelosuppression, followed by acute myeloid leukemia (5.2%). Furthermore, some adverse events were reported in patients with unspecified indications (3.7%), suggesting that some patients might be at risk of myelosuppression due to unclear drug mechanisms or improper treatment choices. Geographically, the majority of DIM reports originated from China (57.4%), followed by Japan (12.4%) and the United States (6.76%). These disparities likely reflect differences in pharmacovigilance systems, reporting practices, and regulatory emphasis rather than true incidence rates. For example, China’s recent advancements in post-marketing surveillance may have contributed to the surge in adverse event reporting. Likewise, the U.S. FDA’s open-access FAERS system facilitates active reporting by healthcare providers and patients. In contrast, regions with limited pharmacovigilance infrastructure or underdeveloped reporting systems may be underrepresented. Additionally, variations in drug usage patterns and patient demographics across countries could further influence the observed distribution. These geographical imbalances introduce potential reporting bias, which may affect the interpretation and generalizability of the findings. Therefore, caution is warranted when extrapolating these results to countries with different healthcare systems or lower reporting capacity.
Based on these analyses, clinical practices should develop personalized medication monitoring plans tailored to the characteristics of different patient populations. For female patients and those aged 18-65, closer monitoring of drug use is essential, especially for drugs with a high risk of myelosuppression (e.g., certain chemotherapy agents). 18 For patients with low body weight or missing weight data, special attention should be given to potential differences in drug metabolism, with regular assessments of blood routine tests to detect potential myelosuppression risks early. 19 Additionally, for patients with breast cancer or acute myeloid leukemia, given the high incidence of myelosuppression in these populations, bone marrow function monitoring should be strengthened during treatment to ensure early detection of abnormalities and timely adjustment of treatment plans. Finally, for patients from different countries or regions, geographical differences in drug safety should be considered, and treatment strategies and monitoring measures should be appropriately adjusted to ensure patient safety and therapeutic efficacy.
Major High-Risk Drug Categories
Based on FAERS data and the ROR algorithm, this study identified 15 high-risk drug signals associated with myelosuppression, mainly involving chemotherapy, targeted therapy, and immunotherapy agents. While a higher ROR suggests a stronger statistical association, it does not imply clinical severity or causality. Currently, no standardized threshold defines a “high-risk” drug based solely on disproportionality metrics, which serve primarily for signal detection. However, drugs showing strong, consistent signals—particularly those without prior labeling for myelosuppression—warrant further validation. If confirmed by real-world studies or clinical trials, these findings could inform regulatory actions such as label revisions, enhanced surveillance, or clinical warnings.
Chemotherapy Drugs (e.g., Cyclophosphamide, Cytarabine, Epirubicin, Eribulin): Chemotherapy drugs primarily induce myelosuppression by directly killing rapidly dividing cells, including hematopoietic stem cells in the bone marrow. Cyclophosphamide, an alkylating agent widely used for the treatment of various cancers, cross-links with DNA to inhibit DNA replication and repair, leading to a reduction in bone marrow hematopoietic function and subsequently causing myelosuppression. 20 Cytarabine, an antimetabolite, inhibits DNA synthesis, directly affecting cell proliferation in the bone marrow, resulting in a reduction of white blood cells, red blood cells, and platelets. 21 Similarly, Epirubicin, an anthracycline anticancer agent, binds to DNA and inhibits its synthesis, leading to widespread myelosuppression. 22 Eribulin, a microtubule inhibitor, prevents tumor cell division and, while mainly targeting cancer cells, also exerts toxic effects on hematopoietic cells in the bone marrow. 23 These drugs require close monitoring of blood routine tests in clinical practice, with timely adjustments to drug doses to prevent severe myelosuppression.
Targeted Therapy Drugs: Targeted therapy drugs act through specific molecular targets and indirectly affect hematopoietic function, leading to myelosuppression. Several drugs identified in this study have well-documented mechanisms contributing to myelosuppression, which align with existing literature. Venetoclax, a BCL-2 inhibitor, induces apoptosis in cancer cells but also disrupts normal hematopoiesis by inhibiting BCL-2 in bone marrow cells, leading to cytopenias. 24 Gilteritinib, used for FLT3-mutated AML, suppresses leukemic proliferation but can also impair normal bone marrow cell growth, increasing the risk of myelosuppression. 25 Abemaciclib, a CDK4/6 inhibitor, inhibits cancer cell cycle progression but concurrently reduces white blood cell and platelet counts by affecting hematopoietic progenitor cells. 26 Trastuzumab emtansine, an antibody-drug conjugate targeting HER2-positive cells, delivers cytotoxic agents that may inadvertently damage bone marrow function, resulting in hematologic toxicity. 27
These findings are consistent with prior pharmacological and clinical reports, supporting the robustness of our signal detection results. However, the identification of additional signals in less-studied drugs—such as pertuzumab*, sodium chloride*, and mesna*—extends the current understanding of drug-induced myelosuppression and highlights areas where further clinical validation is needed.
Immunotherapy Drugs: Immunotherapy drugs exert anticancer effects by modulating the immune system but may sometimes induce immune-mediated myelosuppression or suppress normal hematopoiesis through immune responses. Tislelizumab and Toripalimab, PD-1 inhibitors, enhance T-cell attacks on tumors by relieving immune checkpoint inhibition. However, this immune response may lead to immune-mediated myelosuppression. 28 Daratumumab, which targets CD38, is primarily used in the treatment of multiple myeloma. While it is effective in controlling the tumor, it can interfere with the proliferation of normal bone marrow cells, leading to myelosuppression. 29 Obinutuzumab, a CD20-targeting monoclonal antibody, is commonly used to treat B-cell-related malignancies but may affect bone marrow function by depleting B-cells, leading to myelosuppression. 30
These drugs, through different mechanisms, directly or indirectly influence hematopoiesis in the bone marrow, resulting in a reduction in white blood cells, red blood cells, and platelets, thereby causing myelosuppression. Myelosuppressive reactions to these drugs require strict monitoring and management in clinical practice, including regular blood routine tests, adjustment of drug doses, assessment of potential side effects, and individualized treatment plans based on the patient’s clinical status. By optimizing drug use strategies and enhancing drug safety monitoring, the risk of myelosuppression can be effectively reduced, improving treatment outcomes and the quality of life for patients.
Drug-Induced Onset Time and Characteristics
In this study, we analyzed the onset time of adverse events related to myelosuppression induced by drugs, and the results revealed significant differences in the onset time characteristics across various drugs. For Palbociclib, the median onset time for adverse events was 180 days. The Weibull distribution shape parameter (β < 1, 95% CI < 1) indicates that the incidence of myelosuppression-related adverse events is relatively high in the early stages of treatment and decreases over time, exhibiting a typical “early failure-type curve.” Similar onset time characteristics were observed for Trastuzumab, Bevacizumab, Venetoclax, Methotrexate, and Pertuzumab, which may cause rapid and significant disruption to the immune system or hematopoietic function during the initial phase of treatment, resulting in a higher incidence of adverse events in the short term. 31
The early failure-type curves observed in this study indicate that the risk of myelosuppression is elevated during the initial phase of treatment, particularly within the first 2-4 weeks. This highlights the importance of early monitoring in clinical practice. Frequent blood tests—including white blood cell, red blood cell, and platelet counts—should be performed to promptly detect hematologic toxicity. For high-risk patients, such as those receiving combination therapies or with pre-existing blood disorders, individualized dose adjustments or delayed dose escalation may be warranted. These strategies can facilitate timely intervention and reduce the risk of severe complications.
Limitations
Despite rigorous signal detection, our study has several limitations. While duplicate reports were removed using unique case identifiers, undetected duplicates may still exist. Geographic reporting bias may also influence results, as countries with more active pharmacovigilance systems (e.g., China, Japan) contributed disproportionately to the total reports. Additionally, drug-specific treatment durations were not uniformly available, limiting our ability to assess the time-dependent nature of myelosuppression risk. Finally, although 12 new signals were identified, these findings have not yet been validated in other databases or clinical settings. Further studies using real-world clinical data or prospective cohorts are warranted to confirm these associations and establish causal relationships.
Conclusion
This study conducted an in-depth analysis of drug-induced myelosuppression (DIM) using the FDA Adverse Event Reporting System (FAERS) and identified drug signals associated with DIM, exploring its occurrence characteristics and risk factors. Through the analysis of 21 380 DIM-related adverse event reports, the study revealed that female patients, individuals aged 18 to 65, and breast cancer patients are at higher risk for DIM. Additionally, factors such as weight and geographic distribution were found to influence the occurrence of DIM. The study identified drugs such as trastuzumab, bevacizumab, and venetoclax as being more strongly associated with DIM, with some of these drugs not clearly indicating the risk of myelosuppression in their labeling, and new risk signals were discovered.
Furthermore, through signal mining methods, 12 new drug signals were identified, showing a high signal strength but not labeled with the DIM risk. This research provides critical drug safety data for regulatory agencies, clinicians, and patients, advancing the scientific basis for DIM risk assessment and early detection. By optimizing drug usage monitoring and individualizing treatment plans, the risk of DIM could be reduced, improving treatment outcomes and quality of life for patients. However, the findings still require confirmation through further clinical validation and long-term data analysis.
Supplemental Material
Supplemental Material - Signal Mining and Analysis of Drug-Induced Myelosuppression: A Real-World Study From FAERS
Supplemental Material for Signal Mining and Analysis of Drug-Induced Myelosuppression: A Real-World Study From FAERS by Kaiyue Xia, Shupeng Chen, Yingjian Zeng, Nana Tang, and Meiling Zhang in Cancer Control
Acknowledgments
This study was performed using open-source data provided by the FAERS database, and we thank all those who provided information for this database.
Author contributions: Kaiyue Xia : Methodology, Data curation, Conceptualization. ShuPeng Chen : Writing – original draft. Yingjian Zeng : Methodology, Data curation, Conceptualization. Nana Tang: Methodology, Formal analysis, Data curation. Meiling Zhang : Writing – review & editing. All authors contributed to manuscript revision, and read and approved the submitted version.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work has been supported and funded by the National Natural Science Foundation of China project (82260914), the Jiangxi Provincial Natural Science Foundation project (20192BAB205108), and the Jiangxi Provincial Administration of Traditional Chinese Medicine Traditional Chinese Medicine Advantageous Disease Cultivation Project (Gan Cai She Zhi [2023] No. 70).
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Supplemental Material: Supplemental material for this article is available online.
ORCID iDs
Shupeng Chen https://orcid.org/0009-0007-8064-4840
Yingjian Zeng https://orcid.org/0009-0005-0949-2982
Ethical Statement
This study used publicly available data from the FDA Adverse Event Reporting System (FAERS) and did not involve human participants, human tissue, or identifiable private information. Therefore, ethical approval and informed consent were not required, in accordance with institutional and international guidelines.
Data Availability Statement
The original contributions presented in the study are included in the article/Supplemental Material, further inquiries can be directed to the corresponding author. *
References
1 Epstein RS Aapro MS Basu Roy UK , et al. Patient burden and real-world management of chemotherapy-induced myelosuppression: results from an online survey of patients with solid tumors . Adv Ther . 2020 ; 37 : 3606 - 3618 . doi: 10.1007/s12325-020-01419-6 32642965 PMC7340862
2 Crawford J Herndon D Gmitter K Weiss J . The impact of myelosuppression on quality of life of patients treated with chemotherapy . Future Oncol . 2024 ; 20 : 1515 - 1530 . doi: 10.2217/fon-2023-0513 38587388 PMC11441072
3 Zhang Y Liu J Wang Y , et al. Nucleosides and amino acids, isolated from cordyceps sinensis, protected against cyclophosphamide-induced myelosuppression in mice . Nat Prod Res . 2022 ; 36 : 6056 - 6059 . doi: 10.1080/14786419.2022.2043307 35188001
4 Salem AM Dvergsten E Karovic S Maitland ML Gopalakrishnan M . Model-based approach to identify predictors of paclitaxel-induced myelosuppression in “real-world” administration . CPT Pharmacometrics Syst Pharmacol . 2023 ; 12 : 929 - 940 . doi: 10.1002/psp4.12963 37101403 PMC10349185
5 Fowler-Shorten DJ Hellmich C Markham M Bowles KM Rushworth SA . Bcl-2 inhibition in haematological malignancies: clinical application and complications . Blood Rev . 2024 ; 65 : 101195 . doi: 10.1016/j.blre.2024.101195 38523032
6 Brijs J Peczynski C Boreland W , et al. Immune checkpoint inhibitors therapy for solid organ malignancies after allogeneic hematopoietic stem cell transplantation: a retrospective study from the ebmt transplant complications working party . Bone Marrow Transplant . 2024 ; 60 : 415 - 417 . doi: 10.1038/s41409-024-02497-8 39668190
7 Bosi A Lund LC Mahalingasivam V , et al. Drug use and acute kidney injury: a drug-wide association study (dwas) in Denmark and Sweden . Clin Kidney J . 2025 ; 18 : sfae338 . doi: 10.1093/ckj/sfae338 39802588 PMC11719035
8 Benchimol EI Smeeth L Guttmann A , et al. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement . PLoS Med . 2015 ; 12 ( 10 ): e1001885 . doi: 10.1371/journal.pmed.1001885 26440803 PMC4595218
9 Rothman KJ Lanes S Sacks ST . The reporting odds ratio and its advantages over the proportional reporting ratio . Pharmacoepidemiol Drug Saf . 2004 ; 13 : 519 - 523 . doi: 10.1002/pds.1001 15317031
10 Evans SJ Waller PC Davis S . Use of proportional reporting ratios (prrs) for signal generation from spontaneous adverse drug reaction reports . Pharmacoepidemiol Drug Saf . 2001 ; 10 : 483 - 486 . doi: 10.1002/pds.677 11828828
11 Bate A Lindquist M Edwards IR , et al. A bayesian neural network method for adverse drug reaction signal generation . Eur J Clin Pharmacol . 1998 ; 54 : 315 - 321 . doi: 10.1007/s002280050466 9696956
12 Fan Y Wu T Xu P , et al. Neratinib safety evaluation: real-world adverse event analysis from the faers database . Front Pharmacol . 2024 ; 15 : 1425171 . doi: 10.3389/fphar.2024.1425171 39346561 PMC11427278
13 Alkhaldi H Kharfan-Dabaja M El Fakih R Aljurf M . Safety and efficacy of immune checkpoint inhibitors after allogeneic hematopoietic cell transplantation . Bone Marrow Transplant . 2023 ; 58 : 1075 - 1083 . doi: 10.1038/s41409-023-02073-6 37516808
14 Liu L Yang B Yuan H , et al. Human serum albumin nanoparticles as a carrier of 20(s)-protopanaxadiol via intramuscular injection to alleviate cyclophosphamide-induced myelosuppression . Mol Pharm . 2023 ; 20 : 5125 - 5134 . doi: 10.1021/acs.molpharmaceut.3c00409 37647098
15 Du Y Yang J Yan H Liu YL Chen XP . Gas5 promotes cytarabine induced myelosuppression via inhibition of hematopoietic stem cell differentiation . Toxicol Appl Pharmacol . 2024 ; 483 : 116841 . doi: 10.1016/j.taap.2024.116841 38290668
16 Taylor SJ Duyvestyn JM Dagger SA , et al . Preventing chemotherapy-induced myelosuppression by repurposing the flt3 inhibitor quizartinib . Sci Transl Med . 2017 ; 9 ( 401 ): eaam8060 . doi: 10.1126/scitranslmed.aam8060 28794285
17 Rapoport BL Shannon VR Cooksley T , et al. Pulmonary toxicities associated with the use of immune checkpoint inhibitors: an update from the immuno-oncology subgroup of the neutropenia, infection & myelosuppression study group of the multinational association for supportive care in cancer . Front Pharmacol . 2021 ; 12 : 743582 . doi: 10.3389/fphar.2021.743582 34675810 PMC8523897
18 Gu J Lin Y Wang Y . Case report: nudt15 polymorphism and severe azathioprine-induced myelosuppression in a young Chinese female with systematic lupus erythematosus: a case analysis and literature review . Front Pharmacol . 2023 ; 14 : 1001559 . doi: 10.3389/fphar.2023.1001559 37229272 PMC10203499
19 Zhang T Krekels EHJ Smit C , et al. Drug pharmacokinetics in the obese population: challenging common assumptions on predictors of obesity-related parameter changes[J] . Expert Opin Drug Metab Toxicol . 2022 , 18(10): 657 - 674 . doi: 10.1080/17425255.2022.2132931 36217846
20 Deng J Zhong Y Wu Y , et al. Carnosine attenuates cyclophosphamide-induced bone marrow suppression by reducing oxidative dna damage . Redox Biol . 2018 ; 14 : 1 - 6 . doi: 10.1016/j.redox.2017.08.003 28826042 PMC5565745
21 Pereira-Vieira J Weber DD Silva S , et al. Glucose metabolism as a potential therapeutic target in cytarabine-resistant acute myeloid leukemia . Pharmaceutics . 2024 ; 16 : 442 . doi: 10.3390/pharmaceutics16040442 38675105 PMC11055074
22 Fan C Georgiou KR McKinnon RA Keefe DMK Howe PRC Xian CJ . Combination chemotherapy with cyclophosphamide, epirubicin and 5-fluorouracil causes trabecular bone loss, bone marrow cell depletion and marrow adiposity in female rats . J Bone Miner Metab . 2016 ; 34 : 277 - 290 . doi: 10.1007/s00774-015-0679-x 26056019
23 Oey O Wijaya W Redfern A . Eribulin in breast cancer: current insights and therapeutic perspectives . World J Exp Med . 2024 ; 14 : 92558 . doi: 10.5493/wjem.v14.i2.92558 38948420 PMC11212747
24 Cao Q Wu X Zhang Q , et al. Mechanisms of action of the bcl-2 inhibitor venetoclax in multiple myeloma: a literature review . Front Pharmacol . 2023 ; 14 : 1291920 . doi: 10.3389/fphar.2023.1291920 38026941 PMC10657905
25 Gozzo L Nardo A Brancati S , et al. Severe gastrointestinal toxicity following the use of gilteritinib: a case series and analysis of postmarketing surveillance data . Healthcare (Basel) . 2023 ; 11 : 1479 . doi: 10.3390/healthcare11101479 37239765 PMC10218212
26 Okada Y Chikura S Kimoto T Iijima T . Cdk4/6 inhibitor-induced bone marrow micronuclei might be caused by cell cycle arrest during erythropoiesis . Genes Environ . 2024 ; 46 : 3 . doi: 10.1186/s41021-024-00298-5 38303098 PMC10832093
27 Kim HJ Yoon J Park YH . Long-term hepatobiliary disorder associated with trastuzumab emtansine pharmacovigilance study using the fda adverse event reporting system database . Sci Rep . 2024 ; 14 : 19587 . doi: 10.1038/s41598-024-69614-x 39179667 PMC11343769
28 He J Luo G Liu S , et al. Tislelizumab plus neoadjuvant chemotherapy and concurrent chemoradiotherapy versus neoadjuvant chemotherapy and concurrent chemoradiotherapy for locally advanced nasopharyngeal carcinoma: a retrospective study . Transl Oncol . 2024 ; 48 : 102058 . doi: 10.1016/j.tranon.2024.102058 39079408 PMC11339050
29 Dima D Dower J Comenzo RL Varga C . Evaluating daratumumab in the treatment of multiple myeloma: safety, efficacy and place in therapy . Cancer Manag Res . 2020 ; 12 : 7891 - 7903 . doi: 10.2147/CMAR.S212526 32904669 PMC7457558
30 Masamoto Y Taoka K Maki H Kurokawa M . Bone marrow involvement is a risk factor for infusion-related reactions in patients with follicular lymphoma treated by obinutuzumab . Ann Hematol . 2022 ; 101 : 2795 - 2797 . doi: 10.1007/s00277-022-04987-5 36192661
31 Lamble AJ Kosaka Y Laderas T , et al. Reversible suppression of t cell function in the bone marrow microenvironment of acute myeloid leukemia . Proc Natl Acad Sci U S A . 2020 ; 117 : 14331 - 14341 . doi: 10.1073/pnas.1916206117 32513686 PMC7321988