Potential Drug Interactions Between Psychotropics and Intravenous Chemotherapeutics Used by Patients With Cancer

Abstract

Introduction: Patients undergoing cancer treatment usually have comorbidities, and psychiatric disorders are commonly seen in these patients. For the treatment of these psychiatric disorders, the use of psychotropic drugs is common, turning these patients susceptible to untoward drug interactions. Therefore, the aim of this study was to estimate the prevalence of clinically relevant drug–drug interactions (DDI) between chemotherapeutic and psychotropic agents in patients with cancer treated at an oncology service in southern Brazil. Methods: An observational epidemiological study with a cross-sectional census-type design was carried out between October and December 2020. The drug-drug interactions were identified through consultation and analysis of the Medscape Drug Interaction Check and Micromedex databases. The interactions were classified as major, when the interaction can be fatal and/or require medical intervention to avoid or minimize serious adverse effects and moderate, when the interaction can exacerbate the patient’s condition and/or requires changes in therapy. Results: A total of 74 patients was included in the study among the 194 patients seen in the oncology service during the period studied. A total of 24 (32.4%) DDIs were found, 21 (87.5%) of which were classified as being of major risk and 3 (12.5%) as moderate risk. According to the mechanism of action, 19 (79.1%) were classified as pharmacodynamic interactions and 5 (20.9%) as pharmacokinetic interactions. Conclusion: It was shown that a considerable percentage of patients undergoing intravenous chemotherapy are at risk of pharmacological interaction with psychotropic drugs. Thus, it is essential that the oncologist considers all psychotropic drugs and other drugs used by patients in order to avoid drug-drug interactions.

Introduction

The development of mood disorders are common in patients with cancer,^1,2 occurring in approximately 20% of patients diagnosed with cancer, ² reaching up to 77.4%. ³ Patients diagnosed with neoplasms show more frequently symptoms of depression, hopelessness, fatigue, body image distortions, and social withdrawal.^4-6 For the treatment of such clinical conditions, psychotropic drugs use is frequent,^7,8 however, it is known that the use of these drugs and their combinations can trigger several problems such as untoward drug interactions, adverse reactions and drug tolerance. ⁹

Relating the cancer stage and the development of mood disorders has been the objective of some studies^10-12 and it is suggested that in the diagnostic phase, the development of anxiety symptoms can occur in up to 23.5% of patients, with a progressive reduction of symptoms until the therapeutic phase. In contrast, depression symptoms are observed in up to 14.1% of patients in the diagnostic phase, ¹⁰ but without progression during treatment.^10,11

Data from the French National Health Fund corroborate the relationship between cancer and mood disorders, and reveal that the dispensation of psychotropic drugs to patients with cancer is greater than for patients without cancer, reaching a peak immediately after the diagnosis of the disease; anxiolytics are the most prescribed psychotropic drugs, followed by hypnotics and antidepressants. ⁸ In another study, it was also observed that 14% of patients seen in the Pediatric Oncology sector of the American National Cancer Institute received the prescription of at least one psychotropic drug, with anticonvulsants being the most prescribed class, followed by antidepressants—mainly selective serotonin reuptake inhibitors (SSRIs). ¹³

Due to the different drugs used in chemotherapy and the use of supportive drugs, patients with cancer are more susceptible to the development of untoward drug interaction,^14-16 this being defined as the increase or reduction of the clinical effect of a drug due to the interference of another which can lead to loss of effectiveness or cause toxicity.^17-19 The high prevalence of drug-drug interactions (DDI) in cancer patients has been observed in several studies,^8,14,20-22 that demonstrated that about 4% of deaths among hospitalized patients with cancer are due to such interactions. ²³

Studies carried out in Brazilian hospitals, reviewing the prescriptions of patients who received chemotherapy treatment, found at least one DDI. In addition, the incidence of interactions ranged from 26% to 100% between studies.^20,21 Similarly, data from studies conducted in other countries show DDI rates ranging from 27% to 58% of the patients assessed.^14,22,24 The presence of comorbidities increases the tendency to use a greater number of medications,^25-27 increasing their susceptibility to drug interactions,^20,26,27 notably among antineoplastic and psychotropic drugs.^24,28-30

An expressive DDI between chemotherapeutic and psychotropic drugs has been observed in patients undergoing cancer treatment,^14,31 and the psychotropics most commonly cited in these interactions are antiepileptics, antipsychotics, barbiturates, and benzodiazepines,^24,28,31 more specifically carbamazepine, phenobarbital, primidone, ³¹ haloperidol, alprazolam, ²⁸ sodium valproate,^28,31 and phenytoin.^24,31 Therefore, in order to avoid the negative impacts of DDI in patients with cancer, it is important that early detection of these potential drug interactions occurs in order to avoid their use. In view of the above, the objective of this study was to estimate the prevalence of potential clinically relevant drug interactions between chemotherapeutic and psychotropic agents in patients with cancer treated at an oncology service in southern Brazil.

Methods

An observational epidemiological study with a cross-sectional census-type design was carried out between October and December 2020. The population studied consisted of all patients with cancer who were using psychotropics and who were receiving intravenous chemotherapy treatment in an oncology service in Southern Brazil that is part of the National Clinical Research Network, being a reference in the service of High Complexity in Oncology for the Ministry of Health. Patients receiving intravenous chemotherapy corresponds to the majority of cancer treatments in the sector and therefore were included in the study.

The oncology service in which the research was carried out provides assistance to patients with cancer attended by the Government Unified Health System (SUS) or covered by health insurance and also private patients, providing chemotherapy treatment for all types of solid tumors.

Procedures

Data collection was carried out at 2 different times: First, patients with cancer who were undergoing treatment with psychotropics and who were receiving intravenous chemotherapy in the oncology service were interviewed. Clinical and demographic data were obtained including gender, age, occupation, type of cancer, comorbidities, medication in use, alcohol consumption, smoking, and school education. Patients with less than 10 years of schooling were considered as low educational level.

Polypharmacy was defined as the use of at least 5 medications daily.^32,33 Patients aged 60 years or over were considered elderly. To calculate the length of cancer treatment, the date informed by the patient or the date on which the patient started cancer treatment at the institution was considered.

Sampling was obtained in a nonrandom manner, because of the demand at the service during data collection. In a second phase, patients’ medical records were accessed in order to obtain data on the chemotherapeutic drugs used in oncological treatment, as well as psychotropic drugs and other data contained in the patient’s medical record. Antidepressants, anxiolytics, hypnotics, sedatives, anticonvulsants, and antipsychotics were considered psychotropic medications.

The inclusion criteria established were as follows: to be under treatment with at least one intravenous chemotherapeutic drug and receive treatment with at least one psychotropic drug. Patients who did not accept to participate in the investigation and those for whom the investigators were unable to determine the name of the drugs in use were excluded.

Potential drug interactions were identified through consultation and analysis of databases Medscape Drug Interaction Check ³⁴ and Micromedex2®. ³⁵ The first one classifies interactions according to the mechanism of action, while the second allows classification of the interactions according to the clinical risk and mechanism of action. The interactions were classified as contraindicated, when the concomitant use could result in death; major, when the interaction can be fatal and/or require medical intervention to avoid or minimize serious adverse effects; moderate, when the interaction can exacerbate the patient’s condition and/or requires changes in therapy; and minor, when the interaction is expected to have limited clinical effects. Regarding the mechanism of action, drug interactions were classified as pharmacokinetic and pharmacodynamic DDI. By pharmacokinetics, it is understood that the interaction interferes with the absorption, distribution, metabolism, and/or excretion of the drug. On the other hand, when the interaction results in synergism or antagonism, it is defined as pharmacodynamic interaction.^36,37

Data Analysis

The data collected were entered into the Microsoft Office Excel software (version 16.45) and the statistical analysis was performed using SPSS Statistics® version 26 (IBM, Armonk, New York, USA). Mean, median and standard deviation were calculated for continuous variables and percentages were calculated for categorical variables. Pearson’s chi-square test was used to test the association between the variables of interest and the Student’s t test to compare the means.

An exploratory analysis was carried out to assess the normality of the distribution (Kolmogorov-Smirnov) of the continuous variables. For the variables with normal distribution, a parametric analysis was conducted. The Mann-Whitney test (U) was performed for nonparametric variables. The analysis of the association between the variables of interest was performed using the chi-square test. The level of statistical significance adopted was 5%.

The data obtained and the identification of the research participants were kept confidential, as recommended by Resolution 466/2012 of the National Health Council. The research was approved by the Research Ethics Committee of the University of Southern Santa Catarina on May 10, 2020 (Opinion No. 4 031 982).

Results

Between October and December 2020, 194 patients were seen in the oncology service. A total of 120 patients were excluded from the investigation, because some did not use psychotropic drugs, some did not know the name of the psychotropic drugs used, or because of a divergence between the information in connection with the use of psychotropic drugs indicated by the participants with the information found in the electronic medical records (EMR). Thus, 74 patients were included in the study. These data are reported in Figure 1.

Figure 1.

Flowchart indicating patient exclusion criteria and the pharmacological interactions found.

Patients who did not use psychotropic drugs during the chemotherapy period or when the researchers were unable to access the name of the psychotropic agent used were excluded.

The average age of participants included in the survey was 59 years (SD = 12.1). The tumors diagnosed as the most prevalent primary cancer in the patients surveyed were those of the gastrointestinal tract (39.2%) and breast cancer (29.7%). Most participants (64.9%) had been undergoing chemotherapy for less than a year. The sociodemographic characteristics of the participants included in the study are shown in Table 1.

Table 1.

Sociodemographic Characteristics of Patients (n = 74).

Variables	N	%
Age (years)
20-40	7	9.5
41-60	31	41.9
>60	36	48.6
Gender
Male	25	33.8
Female	49	66.2
Education
High education level a	28	37.8
Low education level b	46	62.2
Marrital status
Married or living with partner	45	60.8
Widowed	12	16.2
Divorced or separated	9	12.2
Never married	8	10.8
Family history of cancer
No	28	37.8
Yes, maternal lineage	23	31.1
Yes, paternal lineage	20	27.0
Yes, paternal and maternal lineage.	3	4.1
Smoking habit
Yes	2	2.7
No	44	59.5
Ex-smoker	28	37.8
Alcohol consumption
Yes	5	6.8
No	69	93.2
Site of the neoplasia
Gastrointestinal	29	39.2
Breast	22	29.7
Genitourinary	9	12.2
Gynecologic	4	5.4
Head and neck cancer	4	5.4
Skin	3	4.1
Lung	2	2.7
Neoplasms of uncertain or unknown behavior	1	1.4
Lenght of cancer treatment (months)
0-12	48	64.9
13-24	8	10.8
25-36	6	8.1
37-48	2	2.7
>48	10	13.6
Other chronic diseases
Yes	49	66.2
No	25	33.8

Values are n (%).

^aEqual or above 10 years of schooling.

^bLess than 10 years of schooling.

Approximately two-thirds of the patients reported the coexistence of other illnesses, in addition to the neoplasia for which they were being treated. Out of the 74 patients, 36 (26.9%) reported having systemic arterial hypertension, 18 (13.4%) had depressive disorders, 16 (11.9%) diabetes mellitus, 12 (9%) anxiety disorders, and 11 (8.2%) had dyslipidemia. In addition to these comorbidities, 10 (7.5%) patients had respiratory disorders, 9 (6.7%) had hypothyroidism, and 9 (6.7%) had osteoarticular diseases.

The patients evaluated in the survey used an average of 4.66 (SD = 3.26) medications per day, with 38 (51.4%) patients consuming daily 2 or more psychotropic drugs. In addition, out of the 74 patients, 58 (78.4%) used psychiatric drugs that were not prescribed by the oncologist responsible for the chemotherapy treatment. The main characteristics of the participants’ pharmacological treatment can be found in Table 2.

Table 2.

Pharmacological Characteristics of Patients (n = 74).

Variables	n	%
Current conventional therapies
Chemotherapy	50	67.6
Chemotherapy + radiotherapy	23	31.1
Chemotherapy + radiotherapy + immunotherapy	1	1.4
Polypharmacy a
Yes	35	47.3
No	39	52.7
Number of psychotropic drugs taken daily
1	36	48.6
2	24	32.4
3	11	14.9
≥4	3	4.1
Psychotropic drugs prescribed by the oncologist
Yes	16	21.6
No	58	78.4
Anticancer drugs mostly involved in interactions
Oxaliplatin	17	70.8
Paclitaxel	3	12.5
Cisplatin	2	8.3
Doxorubicin	1	4.2
Irinotecan	1	4.2
Psychotropic drugs mostly involved in interactions
Antidepressant	17	70.8
Antipsychotic	4	16.7
Anticonvulsant	3	12.5

^aUse of 5 or more medications daily.

The independence test showed that there is no association between clinical, sociodemographic and DDI variables in the study participants. The test results are shown in Table 3.

Table 3.

Clinical and Sociodemographic Variables and their Statistical Associations With Pharmacological Interactions.

Variables	Interaction n (%)		P-value
	Yes	No
Gender
Male	7 (28.0)	18 (72.0)	0.893
Female	13 (26.5)	36 (73.5)
Age (years)
20-40	0 (0.0)	7 (100.0)	0.15
41-60	11 (35.5)	20 (64.5)
>60	9 (25.0)	27 (75.0)
Education
High education level a	9 (32.1)	19 (67.9)	0.439
Low education level b	11 (23.9)	35 (76.1)
Polypharmacy c
Yes	10 (26.5)	29 (74.4)	0.777
No	10 (28.6)	25 (71.4)
Other chronic diseases
Yes	13 (26.5)	36 (73.5)	0.893
No	7 (28.0)	18 (72)
Lenght of cancer treatment (months)
≤12 meses	10 (20.8)	38 (79.2)	0.103
>12 meses	10 (38.5)	16 (61.5)
Values are n (%).
a : ≥60 anos, b : <60 anos

^aEqual or above 10 years of schooling.

^bLess than 10 years of schooling.

^cUse of 5 or more medications daily.

The Mann-Whitney U test showed that the occurrence of DDI between chemotherapy and psychotropic drugs was not associated with age and the duration of cancer treatment; however, the number of psychotropics taken daily was connected to the occurrence of DDI. The median and standard deviation values are shown in Table 4.

Table 4.

Association Between Drug Interaction and Variables.

Variables	Mediana	SD	P-value
Age	60	12.8	0.884
Polypharmacy	4	3.26	0.863
Number of psychotropic drugs taken daily	2	0.86	0.045
Lenght of cancer treatment	8	20.9	0.05

Values are median ± standard deviation (SD).

According to the criteria used in the study, clinically significant interactions between psychotropic and chemotherapeutic agents were identified in 21 (28.3%) patients. Due to the fact that 3 patients had 2 DDIs, a total of 24 interactions were counted. Among the 24 interactions found, 21 (87.5%) were classified as major and 3 (12.5%) as moderate. In addition, interactions were categorized according to the mechanism of action, resulting in 19 (79.1%) pharmacodynamic interactions and 5 (20.9%) pharmacokinetic interactions.

The antineoplastic drugs most frequently associated with interactions with psychotropics were oxaliplatin (n = 17; 70.8%), followed by paclitaxel (n = 3; 12.5%), cisplatin (n = 2; 8.3%), doxorubicin (n = 1; 4.2%), and irinotecan (n = 1; 4.2%). On the other hand, the psychotropic classes with the highest prevalence of interactions with chemotherapeutic drugs were SSRIs (n = 13; 54.2%), followed by antipsychotics (n = 4; 16.7%), anticonvulsants (n = 3; 12.5 %), tricyclic antidepressants (n = 3; 12.5%), and selective serotonin-noradrenaline reuptake inhibitors (SSNRIs) (n = 1; 4.2%).

The main DDIs are attributed to the concomitant use of oxaliplatin and sertraline (16.7%), paclitaxel and carbamazepine (12.5%) and oxaliplatin and amitriptyline (12.5%). The description of DDIs between chemotherapeutic and psychotropic drugs can be seen in Table 5.

Table 5.

Interactions Involving Psycotropic and Chemotherapeutic Drugs (n = 24).

Interactions	Clinical effect	Severity	MA	Level of evidence	N	%
Cisplatin + haloperidol	Increased risk of QT-interval prolongation	Moderate	PD	Satisfactory	2	8.3
Doxorubicin + fluoxetine	Increased doxorubicin levels	Major	PK	Satisfactory	1	4.2
Irinotecan + citalopram	Increased risk of myopathy or rhabdomyolysis	Moderate	PK	Good	1	4.2
Oxaliplatin + amitriptyline	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	3	12.5
Oxaliplatin + citalopram	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	2	8.3
Oxaliplatin + escitalopram	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	2	8.3
Oxaliplatin + fluoxetine	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	1	4.2
Oxaliplatin + paroxetine	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	2	8.3
Oxaliplatin + sertraline	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	4	16.7
Oxaliplatin + quetiapine	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	1	4.2
Oxaliplatin + venlafaxine	Increased risk of QT-interval prolongation	Major	PD	Satisfactory	2	8.3
Paclitaxel + carbamazepine	Decrease in level or effect of paclitaxel	Major	PK	Excellent	3	12.5

Abbreviations: MA, mechanism of action; N, number of prescriptions; PD, pharmacodynamic; PK, pharmacokinetic.

Discussion

The present study shows the prevalence, associated factors and severity of DDI in patients undergoing intravenous chemotherapy and who are on a concomitant use of psychotropic drugs. The frequency of interactions between psychotropic and antineoplastic drugs is shown according to the clinical risk for the patient and the mechanism of action. It was observed that approximately one third of the patients presented a potential risk for interaction between chemotherapeutic drugs and psychotropic drugs. This result corroborates with previous research that estimated between 14% and 75% the prevalence of DDI involving antineoplastic drugs.^13,20,27

Most of the interactions found were classified as pharmacodynamic and previous studies show divergent results in relation to this variable.^14,20,21 Those data are of concern because, although pharmacodynamic interactions have greater potential to be harmful due to the toxic or antagonistic effects, pharmacokinetic interactions can also result in relevant clinical consequences. ³⁸ Thus, patients included in this study are at increased risk of failing therapy with the psychotropic or cancer drug. Another alarming fact is the large number of interactions classified as major, corresponding to almost 90% of the interactions found in our study. Previous studies report between 62% to 70% of IDD classified as major.^39-41 It is possible that more major interactions were found in our study because previous research considered all medications used by patients, while in our study only interactions involving chemotherapeutic and psychotropic were considered. This information is of concern due to the potential of these DDIs to cause serious clinical consequences to patients.

Notoriously, elderly patients are more exposed to polypharmacy^21,42,43 and consequently have greater susceptibility to drug interactions.^21,27 However, the results found do not show any difference between elderly and nonelderly participants. This divergence may be due to the fact that the investigation was restricted to patients, who were undergoing outpatient treatment, excluding hospitalized patients who are presumably more seriously ill patients. Furthermore, compared to other studies,^28,44 the average amount of medicines used daily was lower, even among polypharmacy users.

Most DDIs with chemotherapeutic agents occur through interaction with anticonvulsants,^21,22 and the extended use of anticonvulsants is associated with increased clearance of antileukemic agents, reducing the effectiveness of chemotherapy. ⁴⁵ Similarly, the use of CYP3A4 inducers such as carbamazepine increases the metabolism of paclitaxel,^46,47 suggesting that patients receiving anticonvulsants might have enhanced hepatic clearance of paclitaxel, which could result in reduced efficacy of paclitaxel.^48,49 In addition, the use of fluoxetine, a CYP3A4 inhibitor, increases the cytotoxicity of chemotherapeutic drugs such as doxorubicin.^50,51

Platinum compounds, especially oxaliplatin, are already known for their peripheral neurotoxicity,^52,53 which appears to be mediated by the interaction of oxaliplatin with voltage-gated sodium channels, resulting in alteration in the closing kinetics of axonal sodium channels, ⁵³ predisposing the patient to QT interval prolongation syndrome.^54,55 Therefore, the combined use of oxaliplatin with drugs that can increase the QT-interval (ie, sertraline) should be carefully considered as it may increase toxicity or decreases therapeutics effects.^56,57 Platinum compounds interact with several cytochrome enzymes P450 (CYP), mainly CYP3A4 and CYP2D6. ⁵⁸ These enzymes are important for the biotransformation of several drugs; therefore, the concomitant use of psychotropic drugs that interact with these enzymes, such as sertraline, ⁵⁹ paroxetine and fluoxetine, ⁶⁰ quetiapine, ⁶¹ venlafaxine, ⁶² and escitalopram ⁶³ may result in DDI. There is a difficult in finding studies that explain pharmacokinetically or pharmacodynamically the interactions between chemotherapy and psychotropic drugs, which suggests a bias in this study, given that the interactions were analyzed in renowned database, such as Micromedex® and Medscape®. What is known is that the concomitant use of these pharmacological classes may result in some adverse events for the patient.

Most patients in the present study had the psychotropic drugs prescribed by nononcologist physicians, and some patients could not inform the name of the psychotropic drugs used, and in some cases the psychopharmaceuticals used by the patient were not shown in the hospital records. It is known that efficient communication between psychiatrists, oncologists, and other members of the multidisciplinary team are essential for effective treatment. ⁶⁴ In addition, the EMR used by the institution does not have the DDI detection software and studies show that the use of EMR capable of evaluating and alerting potential DDI is important.^65,66 However, although many medical records have the ability to detect possible drug interactions, they can fail to detect up to 33% of medication errors. ⁶⁷

Because there is no integrated medical record by which both the oncologist and the doctor who prescribes psychotropics can access the list of drugs used by the patient, the communication difficulty between the oncologist and the other doctors who treat the patient’s comorbidities increases, ²⁴ thus boosting the risk of incidence of pharmacological interactions. Likewise, the adverse effects of DDI can be mistakenly attributed as effect of chemotherapy, due to the similarity of symptoms, making early diagnosis difficult. ²⁴

Drug reconciliation is an important tool used to assist in the identification and resolution of unintended DDI, involving a multidisciplinary team composed of doctors, pharmacists and nurses,^68,69 being an important component of the medication management process in order to increase safety in the treatment of each patient. ⁷⁰ Therefore, the pharmacist has a relevant role in the prevention of DDI ⁷¹ and, consequently, in reducing the incidence of adverse events. The identification and signaling of DDI are attributions of the clinical pharmacist,^72,73 a professional who is not present in all hospitals and clinics. In institutions that have a clinical pharmacist, this professional is responsible to discuss the interactions found with a multidisciplinary team, ensuring the safety to the patient and supporting to choose the better pharmacotherapy to the oncologic patient treatment.^73,74

In this study, there are some limitations. For example, only patients undergoing outpatient intravenous chemotherapy, with more preserved functionality than hospitalized patients were included. In addition, the study was carried out in a single center with a small sample size and the design of the study did not allow us to verify if the drug interaction indeed occurred. Similarly, the exclusion of patients who did not know how to name the medications used by them may have resulted in the loss of important information in this study. However, the findings of this study are consistent with previously published studies on polypharmacy in patients with cancer.^40,41,75

Conclusion

The results of this investigation show that a considerable portion of patients undergoing intravenous chemotherapy are at risk of pharmacological interaction with psychotropic drugs. The occurrence of potential DDI between chemotherapy and psychotropic is frequent, with some interactions being more clinically relevant than others. The signaling of these interactions is important, considering the reality of many countries, in which the chemotherapy can sometimes be prescribed and administered by physicians not specialized in oncology. Furthermore, even when prescribed by an oncologist, in hospitals or clinics, the search for drug interactions may not be a concern of the prescribing physician.