Prevalence and predictors of dyspnea in acute coronary syndrome patients receiving ticagrelor: A prospective cohort study in Iran Authors: Maryam Taherkhani, Naimeh Kord, Adineh Taherkhani, Niloofar Shirazi, Arian Tavasol Published: 2026 ABSTRACT The occurrence of dyspnea in patients with acute coronary syndrome (ACS) has always been considered a challenging diagnostic and therapeutic clinical scenario. Ticagrelor, a platelet receptor inhibitor, has beneficial effects in the prevention of ischemic events and mortality, but can cause dyspnea in ACS patients. In this study, we intend to investigate the prevalence of dyspnea in patients with ACS who are under treatment with ticagrelor. This prospective observational cohort study included 200 patients diagnosed with ACS from March 2020 to March 2022 and referred to the Cardiology Department of Modarres Hospital. After prescribing the medicine, the presence and severity of dyspnea were recorded daily according to the modified Borg dyspnea scale during hospitalization. Patients were followed up for 3 months after discharge, and the occurrence of dyspnea was investigated based on the characteristics of the patients. There were no major differences between groups in terms of gender, diabetes, hypertension, or smoking status. Most patients had normal lung auscultation, and the majority presented with unstable angina. Our results showed that 21.5% of patients experienced dyspnea, and 3% of patients stopped using drugs due to the severity of dyspnea. The time of the first attack varied from 1.5 to 36 hours after drug administration, and the maximum intensity of the first attack reached 7. The duration of the second attack varied from 2 to 22 days and was observed with a maximum intensity of 6. The age of patients was significantly higher in people who experienced dyspnea compared to those without dyspnea. Dyspnea is a common side effect in ACS patients receiving ticagrelor. The etiology of new-onset dyspnea in these patients can be complex and challenging to evaluate, as conventional factors did not appear to predict its occurrence. Given the significant therapeutic benefits of ticagrelor, discontinuation should only be considered for persistent and intolerable dyspnea associated with the medication. Further large-scale studies are warranted to better understand its clinical significance and to develop optimal management strategies. 1. Introduction Acute coronary syndrome (ACS) encompasses a range of clinical conditions, from ST-elevation myocardial infarction (STEMI) to non-ST-elevation myocardial infarction (NSTEMI) and unstable angina (UA). It typically arises from the rupture of an atherosclerotic plaque, leading to partial or complete arterial thrombosis that can result in myocardial infarction (MI).[1] Platelet activation plays an essential role in the pathogenesis of atherosclerosis and ACS.[2,3] So, antiplatelet therapy is important in current treatment strategies for patients with ACS to reduce the risk of ischemic complications and improve survival. Oral clopidogrel is the most widely used proven antithrombin in patients with ACS. However, these drugs require hepatic activation by cytochrome P450 enzymes, leading to a delayed onset of action and high inter-individual variability in response. They also irreversibly bind to the P2Y12 receptor and are prone to side effects and drug interactions.[4,5] Ticagrelor is an oral platelet P2Y12 receptor inhibitor that is more potent than clopidogrel in inhibiting platelet aggregation and has been shown to reduce the risk of major cardiovascular adverse events in patients with ACS in the platelet inhibition and patient outcomes studies compared to clopidogrel. Still, dyspnea can occur as a side effect early in the course of ticagrelor treatment.[6,7] The occurrence of dyspnea in patients with ACS has always been considered a challenging diagnostic and therapeutic clinical scenario, which has been discussed by various studies.[8–11] Dyspnea is one of the most common and uncomfortable symptoms experienced by patients. It can be caused by various conditions, such as heart, lung, kidney, and liver diseases, as well as anemia and metabolic abnormalities. A significant proportion of ACS patients (at least 25%) may present with dyspnea as the predominant symptom.[12] In addition, ACS patients may develop dyspnea during hospitalization or in the following weeks due to the development of heart failure, lung infection, adverse reaction to beta-blockers, recurrent ischemia, anemia, or other possible complications. Patients with ACS who present with dyspnea as their primary symptom are less likely to be considered patients with a coronary event, therefore less likely to receive proper therapies, and are more likely to have poor outcomes.[13] Dyspnea is a very common side effect of ticagrelor and occurs in a dose-dependent manner with an incidence of 10% to 20% compared to 0% to 6.4% in patients treated with clopidogrel.[14–17] Premature discontinuation of ticagrelor or switching to another antiplatelet agent occurs in 20% of patients, with dyspnea as the main cause. Although premature discontinuation of ticagrelor is rarely associated with adverse cardiovascular outcomes, this phenomenon warrants careful investigation.[18] The most recent European Society of Cardiology guidelines recommend ticagrelor in combination with aspirin as first-line treatment for patients with ACS regardless of initial treatment strategy.[19] While ticagrelor addresses many of clopidogrel’s limitations by providing a faster onset and stronger, more consistent inhibition of platelet activity,[11] the influence of ticagrelor-induced dyspnea on treatment discontinuation and adherence in ACS patients remains insufficiently investigated, despite its possible effect on clinical outcomes. Since the correct evaluation and management of ACS patients with dyspnea is important, this study aimed to investigate the incidence of dyspnea and its effect on the discontinuation of ticagrelor in the first month after percutaneous coronary intervention (PCI) in ACS patients. 2. Material and methods In this prospective observational cohort study, the 200 participants were selected in a nonrandom and consecutive manner from among the patients who were diagnosed with ACS from March 1, 2020, to March 1, 2022, and referred to several hospitals of Shahid Beheshti University of Medical Sciences, Tehran, Iran. Patients without underlying respiratory diseases, heart failure, active bleeding, and aged <80 years were included in the investigation. Patients with left ventricular ejection fraction <35%, those who did not respond to follow-ups, and those who lacked consent to continue participating were excluded from the study. All participants signed informed consent, and the study protocol was approved by the Shahid Beheshti Medical University Ethics Committee (IR.SBMU.MSP.REC.1400.402). Then, the demographic and clinical information of the patients was recorded from their medical history. Patients received 90 mg of ticagrelor twice daily and were visited every day during hospitalization and asked about the complaint of dyspnea; in case of shortness of breath, we asked the patient to rate it from 1 to 10 according to the modified Borg dyspnea scale.[20,21] Objective pulmonary or cardiac investigations (e.g., spirometry, echocardiography at symptom onset) were not routinely performed at the time of dyspnea assessment. Patients who complained of dyspnea with a score above 3 (moderate) were again auscultated. The drug was discontinued in patients if the patient’s dyspnea was more than moderate or was unbearable for the patient. The process of evaluating the presence and severity of dyspnea was continued for 3 months after discharge. Then, demographic and clinical characteristics of the patients were compared in patients with or without dyspnea. Renal function indices (e.g., serum creatinine, estimated glomerular filtration rate), body mass index, detailed pulmonary function testing, and concomitant medications known to influence dyspnea (e.g., beta-blockers, angiotensin-converting enzyme inhibitors) were not systematically collected and therefore could not be included in the regression models. The sample size of 200 patients was determined based on an estimated dyspnea prevalence of approximately 20% among ACS patients receiving ticagrelor, with a 95% confidence level, 5% margin of error, and accounting for potential follow-up losses. Quantitative data were expressed as mean ± standard deviation, while qualitative data were reported as frequencies and percentages. Before analysis, the normality of quantitative variables (e.g., age, peripheral capillary oxygen saturation [SpO2]) was assessed using the Shapiro–Wilk test. A P value <.05 indicated a nonnormal distribution, in which case the Mann–Whitney U test was used instead of a t test. The comparison of age and SpO2 between 2 groups with and without dyspnea was performed using a t test, and the comparison of other variables between the 2 groups was conducted using a chi-square test. To identify independent factors associated with the development of ticagrelor-induced dyspnea, univariate and multivariate logistic regression analyses were performed. We tested for multicollinearity among the independent variables using the variance inflation factor (VIF). All VIF values were <2.0, confirming the absence of multicollinearity in the model. Statistical analyses were conducted using SPSS version 24, and graphs were generated with GraphPad Prism version 8.4.2 (GraphPad Software Inc.). A P value <.05 was regarded as statistically significant. 3. Results 3.1. Basic and clinical characteristics The characteristics of the study group, including demographic, clinical, cardiopulmonary, and dyspnea-related data, are presented in Table 1. This table summarizes the baseline distribution of factors, such as age, sex, SpO2, comorbidities, MI type, and presence or severity of dyspnea during hospitalization and follow-up. Table 1Baseline demographic, clinical, cardiopulmonary, and dyspnea characteristics of the study population (N = 200).ParameterValue (mean ± SD/N [%])*Age (yr)59.47 ± 11.74Sex Male130 (65%) Female70 (35%)SpO2 (%)97.9 ± 1.59Comorbidities Diabetes67 (33.5%) Hypertension74 (37%) Smoking19 (9.5%)Coughing Yes65 (32.5%) No135 (67.5%)Auscultation findings Normal193 (96.5%) Wheezes4 (2%) Rhonchi2 (1%) Coarse crackles1 (0.5%)Type of MI UA156 (78%) Anterior MI21 (10.5%) Inferior MI10 (5%) NSTEMI9 (4.5%) Lateral MI4 (2%)Dyspnea (overall prevalence)43 (21.5%)Time to first dyspnea attack (hours after ticagrelor)3.56 ± 5.09 (range: 1.5–36)Time to second dyspnea attack (days after ticagrelor)4.64 ± 5.34 (range: 2–22)Duration of dyspnea episode (d)2.17 ± 1.63 (range: 1–7)MI = myocardial infarction, NSTEMI = non-ST-elevation myocardial infarction, SD = standard deviation, SpO2 = oxygen saturation, UA = unstable angina.*Qualitative variables are listed as numbers (percentage), and quantitative are listed as mean ± SD. Baseline demographic, clinical, cardiopulmonary, and dyspnea characteristics of the study population (N = 200). MI = myocardial infarction, NSTEMI = non-ST-elevation myocardial infarction, SD = standard deviation, SpO2 = oxygen saturation, UA = unstable angina. Qualitative variables are listed as numbers (percentage), and quantitative are listed as mean ± SD. In our study, 200 ACS patients with a mean age of 59.47 ± 11.74 years were included. The patients were predominantly male. Diabetes and hypertension were the most common comorbidities. Most patients had normal auscultation findings. UA was the most frequent ACS type. The mean SpO2 was 97.9 ± 1.59%. Dyspnea occurred in 21.5% of patients, and ticagrelor was discontinued in 3% due to dyspnea severity. The occurrence time of the first attack varied from 1.5 to 36 hours after drug administration, and the maximum intensity of the first attack reached 7. 3.2. Severity and timing of dyspnea episodes Overall, 43 patients (21.5%) experienced at least 1 episode of dyspnea. Of these 43 patients, 17 patients reported a second attack during the 3-month follow-up period. The severity of the first and subsequent episodes, based on the modified Borg scale, is detailed in Table 2. It should be noted that most of the dyspnea attacks were mild. In the first episode, 78.5% reported no dyspnea, while 1.5% experienced severe and 0.5% very severe symptoms. During the second attack, dyspnea was less common and less intense. Table 2Severity of dyspnea in the first and second attacks in patients according to the modified Borg dyspnea criteria.Severity levelFirst episode (N = 43)Second episode (N = 17)n (% of all patients)n (% of all patients)Very very slight8 (4%)8 (4%)Very slight15 (7.5%)6 (3%)Slight9 (4.5%)1 (0.5%)Moderate6 (3%)–Somewhat severe1 (0.5%)–Severe3 (1.5%)2 (1%)Very severe1 (0.5%)–Very very severe––Maximal––Borg scale – modified Borg dyspnea scale (0 = none, 1 = very slight, 10 = maximal). Severity of dyspnea in the first and second attacks in patients according to the modified Borg dyspnea criteria. Borg scale – modified Borg dyspnea scale (0 = none, 1 = very slight, 10 = maximal). 3.3. Comparison between patients with and without dyspnea The comparison of demographic and clinical variables between patients who developed dyspnea and those who did not is shown in Table 3. Compared with those without dyspnea, affected patients were significantly older, while other demographic and clinical variables (including gender, auscultation findings, cough, diabetes, hypertension, smoking status, MI type, or SpO2) did not differ meaningfully between groups (Table 3). Although not statistically significant, the difference in auscultation findings between the 2 groups showed a trend toward association (P = .08). Patients who did not develop dyspnea were more likely to have normal auscultation results compared to those who experienced dyspnea. Table 3Comparison of study variables between patients with and without dyspnea.ParameterWithout dyspnea (N = 157)With dyspnea (N = 43)P valueGender.99 Male102 (65%)28 (65.1%) Female55 (35%)15 (34.9%)Auscultation.08 Normal154 (98.1%)39 (90.7%) Wheezes2 (1.3%)2 (4.7%) Rhonchi1 (0.6%)1 (2.3%) Crackles0 (0%)1 (2.3%)Coughing.85 Yes52 (33.1%)13 (30.2%) No105 (66.9%)30 (69.8%)Diabetes mellitus.99 Yes53 (33.8%)14 (32.6%) No104 (66.2%)29 (67.4%)Hypertension.99 Yes58 (36.9%)16 (37.2%) No99 (63.1%)27 (62.8%)Smoking.56 Yes14 (8.9%)5 (11.6%) No143 (91.1%)38 (88.4%)Type of MI.60 Anterior MI16 (10.2%)5 (11.6%) Inferior MI9 (5.7%)1 (2.3%) UA122 (77.7%)34 (79.1%) Lateral MI4 (2.5%)0 (0%) NSTEMI6 (3.8%)3 (7%)Age (yr)57.27 ± 10.5767.48 ± 12.41<.0001SpO2 (%)97.95 ± 1.6297.72 ± 1.46.39Qualitative variables are listed as numbers (percentage), and quantitative variables are listed as mean ± SD.MI = myocardial infarction, NSTEMI = non-ST-elevation myocardial infarction, SD = standard deviation, SpO2 = oxygen saturation, UA = unstable angina. Comparison of study variables between patients with and without dyspnea. Qualitative variables are listed as numbers (percentage), and quantitative variables are listed as mean ± SD. MI = myocardial infarction, NSTEMI = non-ST-elevation myocardial infarction, SD = standard deviation, SpO2 = oxygen saturation, UA = unstable angina. 3.4. Predictors of dyspnea In the initial univariate analysis, only 2 factors showed a significant or near-significant association with experiencing dyspnea. Older age showed a statistically significant association with dyspnea (P < .001), and abnormal auscultation findings showed a trend towards significance (P = .08). All other demographic and clinical variables, including gender, diabetes, hypertension, and smoking status, were not significantly associated with the outcome. After adjusting for potential confounders in a multivariate logistic regression model, age remained the only statistically significant independent variable associated with dyspnea. For each 1-year increase in a patient’s age, the odds of experiencing dyspnea increased by 8% (odds ratio = 1.08, 95% confidence interval = 1.04–1.13, P < .001). The marginal association with abnormal auscultation findings seen in the univariate analysis became nonsignificant after controlling for age, suggesting it was not an independent factor. No other variables demonstrated a significant association with the development of dyspnea. Multicollinearity among independent variables was assessed using the VIF, and no evidence of multicollinearity was detected (all VIF values <2.0). The key results of the final multivariate model are summarized in Table 4. Table 4Multivariate logistic regression analysis of risk factors for dyspnea.VariableOR95% CIP valueAge (per yr)1.081.04–1.13<.001Abnormal auscultation1.950.89–4.25.09Diabetes mellitus1.030.51–2.08.94Hypertension1.010.50–2.03.97Male gender0.980.49–1.96.95CI = confidence interval, OR = odds ratio. Multivariate logistic regression analysis of risk factors for dyspnea. CI = confidence interval, OR = odds ratio. 4. Discussion The percentage of reporting dyspnea and discontinuation of medication has been different in various studies. Giannoni et al observed a dyspnea rate of 43%, of which 23% were severe.[22] In another study by Tamakauskas and colleagues, dyspnea with different severities was reported in 34.3% of patients. This study selected patients unlikely to have dyspnea due to other causes. The severity of dyspnea was assessed by the modified Borg acute dyspnea scale, which was used in early clinical trials to investigate the effect of ticagrelor.[23] In the present study, although we tried to select patients who did not have dyspnea due to underlying reasons, the possibility of other reasons leading to dyspnea due to the high-risk nature of patients should be considered. In the platelet inhibition and patient outcomes trial, ticagrelor was associated with a significantly higher incidence of dyspnea than clopidogrel. This early side effect may have contributed to the higher discontinuation rate observed in the ticagrelor group. They showed that the incidence rate of dyspnea in patients who received ticagrelor was 13.8%; of them, about 0.9% discontinued receiving the drug due to persistent dyspnea.[11] Unlike them, our study showed an incidence rate of 21.5% with a medication discontinuation rate of 3%. These differences might reflect population heterogeneity or study design variations. Ticagrelor-dependent dyspnea mainly arises from 2 proposed mechanisms. The most widely cited involves an increase in extracellular adenosine levels due to ticagrelor’s blockade of the equilibrative nucleoside transporter (ENT1/2). By inhibiting adenosine reuptake into cells, ticagrelor prolongs and raises the tissue concentration of adenosine, which then activates adenosine receptors – particularly on vagal C fibers in the bronchial wall. This stimulation can trigger the sensation of breathlessness in some patients. Adenosine may also contribute by causing mild bronchoconstriction or stimulating other airway sensory nerves, further amplifying the dyspnea response.[24,25] A second hypothesis suggests that ticagrelor’s inhibition of P2Y12 receptors on sensory neurons can directly heighten the perception of dyspnea, independently of adenosine effects. Other reversible P2Y12 inhibitors, like cangrelor and elinogrel, have been observed to increase the incidence of dyspnea, supporting this neural mechanism.[23] These mechanisms are generally transient and dose-dependent, explaining why most dyspnea episodes are mild and self-limited. A study similar to our study, from the perspective of patient composition, was followed up over 3 months using a standard protocol that included attendance at the outpatient clinic, phone calls, and file review. They showed that at 3 months of follow-up, 15 patients (14%) had dyspnea, which was determined by the investigator to be related to ticagrelor. All ticagrelor-related dyspnea cases occurred within the first week after initiation, and 10 (67%) occurred within 24 hours of starting ticagrelor administration. Almost all cases were temporary and mild in severity. Some cases lasted 24 hours, while others continued throughout the study.[26] Regardless of the reason for dyspnea in patients taking ticagrelor, studies in the field of discontinuation of the drug due to severe dyspnea have reported almost similar and small values, which indicates the importance of proper management of patients to continue the drug according to its benefits. In comparing 2 groups with dyspnea and without dyspnea, from the perspective of lung auscultation findings, there were significant differences in the age of patients. Moreover, the logistic regression analysis showed that older age appears to be independently associated with dyspnea and with dyspnea-related ticagrelor discontinuation, without implying a causal relationship. For each 1-year increase in a patient’s age, the odds of experiencing dyspnea increased by 8%. Given the observational design and limited covariate adjustment, causality cannot be established. Based on our results, abnormal auscultation findings showed a trend towards significance. In a study by Angiolillo et al,[25] it was elucidated that age is the strongest predictor of dyspnea-related ticagrelor discontinuation. Patients aged 65 to 74 have a 1.7 times higher risk, while those aged 75 or older have a 2.25 times higher risk of stopping ticagrelor due to dyspnea. Other independent predictors of ticagrelor-related dyspnea and discontinuation include current smoking, prior PCI, hypercholesterolemia, previous coronary artery bypass grafting, peripheral artery disease, and obesity (BMI ≥ 30 kg/m2). In line with the results of our study, a multicenter prospective observational study was conducted on hospitalized STEMI or ACS patients with a dose of 180 mg of ticagrelor, and the treatment with ticagrelor was recorded during 1 month of follow-up. Most of their patients were male (75%), 61 ± 10 years. Clinical manifestations were STEMI in 31.1%, NSTEMI in 48.4%, and UA in 20.5% of cases. During the 1-month follow-up, 37 patients (22.6%) complained of dyspnea. Overall, ticagrelor was discontinued in 27 patients (16.7%). The reason for discontinuing ticagrelor was dyspnea in more than 55% of patients. Seven patients (4.2%) were rehospitalized due to dyspnea during follow-up.[18] It seems that differences between the studies have been shown based on the composition of the patients and the studied population. However, severe attacks are generally rare. However, the advantages of the drug and the disadvantages of dyspnea in ACS patients should be considered. Based on the large-scale ONSET/OFFSET clinical trial results, dyspnea was recorded in 38.6%, 9.3%, and 8.3% of patients receiving ticagrelor, clopidogrel, and prasugrel, respectively. In this ONSET/OFFSET study, only 18% of dyspnea episodes in patients treated with ticagrelor were moderate, and no severe dyspnea was reported, and dyspnea was stable in only 3 patients (14%) who discontinued the medication.[27] Another study aimed to compare the antiplatelet treatment patterns of ticagrelor or prasugrel among patients with ACS treated with PCI. After matching the groups, the results showed that readmission was more common in patients using ticagrelor, mainly due to congestive heart failure and premature readmission. The overall costs of the 2 drugs were almost similar, but the total costs related to dyspnea were significantly higher with ticagrelor. Drug switching was also more common in the case of ticagrelor than prasugrel. Although the differences were small, patients receiving prasugrel were likely to continue use for longer than those receiving ticagrelor.[28] Other studies showed no significant difference in long-term ischemic outcomes between prasugrel and ticagrelor.[29,30] Several investigations also showed no significant difference in non-coronary artery bypass grafting-related bleeding between ticagrelor and prasugrel.[31–34] Therefore, based on studies, ticagrelor is more effective than clopidogrel in preventing MI and death after ACS.[35,36] However, dyspnea is a relatively frequent side effect of this newer antiplatelet agent. Dyspnea associated with ticagrelor has been reported to be mild to moderate, rarely leading to drug discontinuation.[37] In a comprehensive multicenter study, Angiolillo et al demonstrated that dyspnea was the predominant cause of ticagrelor discontinuation following PCI, underscoring the clinical challenge of managing this adverse effect to maintain optimal antiplatelet therapy.[25] On the other hand, some cases of hospitalization due to stent thrombosis after switching to ticagrelor due to drug-related dyspnea have been reported.[38,39] There are several limitations in this study that should be considered. First, the nonrandom and consecutive selection of participants from hospitals may limit the generalizability of the results to other populations or healthcare settings. Second, this study relied on patient-reported dyspnea severity using the modified Borg dyspnea scale, which, although a validated tool, introduces a subjective element to the assessment. In addition, the 3-month follow-up period after discharge may have influenced the detection rate and severity of dyspnea episodes. To reduce baseline confounding, patients with known underlying respiratory diseases or heart failure were excluded; however, this may also limit the applicability of the findings to higher-risk clinical populations. Although multivariable logistic regression was performed, the relatively small number of dyspnea events limited the number of covariates that could be reliably included, and the findings should therefore be interpreted with caution. Furthermore, because dyspnea was assessed without concurrent objective pulmonary or cardiac testing, attribution of symptoms specifically to ticagrelor remains uncertain. Finally, residual confounding is likely, as several clinically relevant variables – including renal function, body mass index, baseline pulmonary disease severity, and concomitant medications – were not measured or adjusted for in the analysis. Our study has several strengths that enhance the validity and clinical relevance of its findings. This is a prospective observational cohort design. The design of our study allowed us to systematically collect data on the severity and rate of dyspnea over time. These data provide valuable insights into the real-world incidence and characteristics of ticagrelor-related dyspnea in a specific patient population. The consistent daily assessment of dyspnea during hospitalization and the extended 3-month follow-up period after discharge ensured comprehensive monitoring of this adverse effect. Furthermore, the use of the modified Borg dyspnea scale provided a standardized method for quantifying dyspnea severity, contributing to the objective evaluation of symptoms. In the present study, we observed that some patients experienced dyspnea in the first days of taking ticagrelor, but this side effect rarely led to discontinuation of the drug. Our results suggest that patient and physician education may be important in reducing unnecessary antiplatelet discontinuation in this clinical scenario. Dyspnea was mild in most cases and often disappeared without the need to stop the medication. More extensive studies are needed to clarify this drug’s clinical significance and optimal management strategy. 5. Conclusion Dyspnea is a common side effect in ACS patients treated with ticagrelor, observed in 21.5% of our study cohort. However, only 3% of patients discontinued the drug due to the severity of dyspnea. While the etiology of new-onset dyspnea in these patients can be complex and challenging to evaluate, conventional factors such as cough and lung auscultation findings do not appear to predict its occurrence. These findings should be interpreted in the context of residual confounding and subjective symptom assessment. Given the significant therapeutic benefits of ticagrelor in preventing ischemic events and mortality, drug discontinuation should only be considered in cases of persistent and intolerable dyspnea associated with the medication. Future large-scale studies are warranted to further elucidate the clinical significance of ticagrelor-related dyspnea and to develop optimal management strategies. Clinically, early recognition of ticagrelor-associated dyspnea and appropriate patient counseling may reduce unnecessary discontinuation of an otherwise prognostically beneficial therapy. Author contributions Conceptualization: Maryam Taherkhani, Arian Tavasol. Data curation: Maryam Taherkhani, Naimeh Kord. Methodology: Maryam Taherkhani, Naimeh Kord. Writing – original draft: Maryam Taherkhani, Naimeh Kord, Adineh Taherkhani, Niloofar Shirazi, Arian Tavasol. Writing – review & editing: Maryam Taherkhani, Naimeh Kord, Adineh Taherkhani, Niloofar Shirazi, Arian Tavasol. Formal analysis: Naimeh Kord, Arian Tavasol. Investigation: Adineh Taherkhani. Resources: Adineh Taherkhani. Visualization: Niloofar Shirazi. Project administration: Maryam Taherkhani, Arian Tavasol. Software: Arian Tavasol. Supervision: Maryam Taherkhani, Arian Tavasol. Validation: Arian Tavasol.