Cancer Treatment and Research Communications 26 (2021) 100277

Contents lists available at ScienceDirect

Cancer Treatment and Research Communications

journal homepage: www.sciencedirect.com/journal/cancer-treatment-and-research-communications

Prevalence, severity, and nature of risk factors associated with drug-drug

interactions in geriatric patients receiving cancer chemotherapy: A
prospective study in a tertiary care teaching hospital
Saravana Kumar Ramasubbu a, Sumit Kumar Mahato a, Akash Agnihotri a,
Rajesh Kumar Pasricha c, Uttam Kumar Nath d, Biswadeep Das Dr. a, b, *
a
Department of Pharmacology, All India Institute of Medical Sciences(AIIMS), Virbhadra Road, Rishikesh-249 203, Uttarakhand, India
b
Additional Professor, Department of Pharmacology, All India Institute of Medical Sciences(AIIMS), Virbhadra Road, Rishikesh-249 203, Uttarakhand, India
c
Department of Radiation-Oncology, All India Institute of Medical Sciences(AIIMS), Virbhadra Road, Rishikesh-249 203, Uttarakhand, India
d
Department of Medical-Oncology/Hematology, All India Institute of Medical Sciences(AIIMS), Virbhadra Road, Rishikesh-249 203, Uttarakhand, India

A R T I C L E I N F O A B S T R A C T

Keywords: Introduction: Polypharmacy increases hazard of drug-drug interactions(DDIs), hospitalization, treatment toxicity,
Risk factors;Drug-drug interactions;Geriatric and mortality in elderly individuals with cancer. The present study explores and analyzes prevalence and severity
cancer patients;Anticancer chemotherapy;Pro­ of DDIs in geriatric cancer patients subjected to anticancer chemotherapy, their mechanisms, stratification of
spective observational study;Teaching hospital
severity, and correlation between DDI risk and number of medications taken.
Methods: This was a cross-sectional study conducted between January-July 2019 at the Medical Oncology/He­
matology and Radiation-Oncology Departments, All India Institute of Medical Sciences(AIIMS) Rishikesh. The
study included a convenience sampling of 126 geriatric cancer patients.
Results: 126 patients were enrolled in present study. DDIs were identified in 97.6% of elderly cancer patients, and
88.9% had at least one DDI with antineoplastic medications. Highest number of DDIs involving antineoplastic
medications in any given patient was 12. DDIs involving medications used for treatment of non-cancerous dis­
eases were observed in 83.3% of patients; highest number of interactions identified in any given patient was 15.
Out of 473 interactions, 237(50.1%) DDIs were attributable to pharmacodynamic mechanisms of action. 126
(27%) of DDIs involved pharmacokinetic mechanisms and 110(23.6%) involved unknown mechanisms. In this
present study, total number of DDIs could be positively correlated with total number of medications and number
of health problems.
Conclusions: Geriatric cancer patients are at high risk of DDIs ascribable to polypharmacy. Physicians may utilize
online DDI checking softwares to alert themselves, characterize potential DDIs, and modify medications judi­
ciously. An integrative and algorithmic approach with inclusion of geriatricians, oncologists, cardiologists,
general practitioners, and clinical pharmacologists/ pharmacists is imperative to optimize drug therapy.

Introduction trends from the US and UK showing a gradual increase in drug use in the
elderly [4,5]. Polypharmacy attains more importance in the geriatric
Prescribing medication to elderly patients is complex and fraught age group receiving anticancer chemotherapy making them vulnerable
with multiple challenges. As the general population ages, the number of to adverse drug effects when compared to the elderly without cancer.
older patients (≥ 65 years) in hospitals increases [1,2]. Among elderly There may also be an increment in the hazard of drug-drug interactions
people, age-associated physiological changes lead to many ailments (DDIs), hospitalization, treatment toxicity, and mortality in such pa­
forcing them to resort to taking multiple drugs at a time, i.e., poly­ tients [3,6]. Prevalence of DDIs may be high in the elderly population
pharmacy. Polypharmacy is defined as the concurrent use of five or more who are undergoing chemotherapy, especially among the newly diag­
medications and has emerged as a significant public health issue nosed patients. There are reports in the medical literature that elderly
amongst an increasingly aging population [3]. This is also reflected in patients documented 33–69% major DDIs arising out of additional

* Corresponding author: Additional Professor, Department of Pharmacology, All India Institute of Medical Sciences(AIIMS), Virbhadra Road, Rishikesh-249 203,
Uttarakhand, India

https://doi.org/10.1016/j.ctarc.2020.100277

Available online 11 December 2020

2468-2942/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

medications and chemotherapeutic agents with a narrow therapeutic have not been incorporated as they are part of the same treatment
window [7,8]. The hazard of drug-related toxicity in geriatric subjects is protocol. The mechanism of DDIs has been classified as “Pharmacoki­
likely to be accentuated with (1) administration or consumption of an netic”, “Pharmacodynamic” and “Unknown”.
increasing number of medications, (2) many prescribers for a single “Medscape Drug Interactions Checker” stratifies the severity of DDIs
patient, (3) the administration/self-administration of over-the-counter into “Contraindicated”, “Serious-use alternative”, “Monitor Closely” and
(OTC) agents, herbs, and/or nutraceuticals with/without an under­ “Minor”.
standing of prescriber/patient, (4) comorbidities and organ functioning The drug regimen for each patient was also scrutinized for risk
compromise, and (5) the presence of certain types of tumors. stratification of drug combinations predisposing to QT-interval pro­
DDIs have been defined as the changes in the effects of one drug due traction/Torsades de pointes (TdP) using the Arizona Center for Edu­
to the presence of another drug in the same body system. DDIs were cation and Research on Therapeutics (AzCERT) Lists (latest). The hazard
responsible for 4.8% of admissions among elderly people and 17% of the stratifications used are “Known Risk”(List 1), “Possible Risk”(List 2),
ADRs causing hospitalizations were attributable to DDIs [9,10]. Many “Conditional Risk”(List 3), and “Drugs to avoid in congenital long QT
hospitalizations caused by DDIs are avoidable. DDIs may become clini­ syndrome”(List 4).
cally relevant with medications bearing a narrow therapeutic index,
medications undergoing zero-order pharmacokinetics, drugs exhibiting Statistical analysis
microsomal enzyme inhibition or induction, and in terminally afflicted
patients with hepatic and/or renal impairment. There is sparse data in Descriptive statistics were used to compute patients’ demographics,
India concerning DDIs in geriatric cancer patients who have been pre­ tumor types, comorbidities, and characteristics of drug interactions.
scribed anticancer chemotherapy. The present study aims to explore and Pearson correlation test was applied to estimate the correlation between
analyze the prevalence and severity of DDIs in geriatric cancer patients age, number of drugs, number of health problems, and number of drug
subjected to anticancer chemotherapy, their mechanisms, stratification interactions. Binary logistic regression was employed to identify the risk
of severity, and correlation between DDI risk and the number of medi­ factor for the occurrence of DDIs involving anticancer medications.
cations taken by each patient.
Results
Materials and methods
A total of 126 patients were enrolled in the present study in which
Study design and patients males (65.8%) comprised the majority of the patients. The mean age of
participants was 66.6 ± 5.3 years. Solid tumors (80.9%) were encoun­
This was a prospective, cross-sectional study conducted between tered more as compared to hematological malignancies (19.1%). Among
January-July 2019 at the Medical Oncology and Hematology Depart­ the solid tumors, lung cancer (24.6%) followed by oropharyngeal cancer
ment, All India Institute of Medical Sciences (AIIMS) Rishikesh. The (13.4%) and breast cancer (7.9%) were the most common afflictions.
study included a convenience sampling of 126 geriatric cancer patients. Multiple myeloma (9.5%) followed by lymphomas (7.4%) were the most
Convenience sampling is a type of nonprobability or nonrandom sam­ common hematological malignancies. The most common comorbidities
pling where members of the target population that meet certain prac­ associated with participants were hypertension (28.5%) followed by
tical criteria, such as easy accessibility, geographical proximity, diabetes (20.6%) and arthritis (14.2%) (Table 1).
availability at a given time, or the willingness to participate are included The 3 online drug interaction softwares, Drugs.com, Epocrates
for the purpose of the study. Convenience sampling is affordable, easy Interaction Check, and Medscape Drug Interactions Checker, were used
and the subjects are readily available [11]. However, convenience to identify the drug interactions in elderly cancer patients in our study.
sampling is not without limitations; the results from the data have less Sometimes, the interaction between two drugs may not be identified by
clear generalizability [12]. This study was duly approved by the Insti­ one software, other two softwares were then used to identify the in­
tutional Ethics Committee (IEC)(AIIMS/IEC/18/161 dated 4.1.2018) teractions. Thus, if any drug interaction was found by at least one (≥ 1)
and written informed consent was taken from all participants. software, it was included in the analysis as a DDI.
Patients (age > 60 years) who had been diagnosed with cancer (solid Out of 126 patients, 123 had DDIs. DDIs in 123 out of 126(97.6%)
tumor or hematological malignancy) and under treatment with ≥ 1 elderly cancer patients are inclusive of the drug interactions identified in
anticancer medication were included in the study. the pharmacotherapy of both cancer and non-neoplastic conditions/
A structured customized proforma was employed to tap socio­ diseases. Out of 126 patients, 112(88.9%) had DDIs involving anti-
demographic and pharmaco-therapeutic data related to this study. cancer drugs. 11 out of 126(8.7%) patients had DDIs only between the
Clinical data pertinent to cancer treatment and other health problems drugs used to treat non-neoplastic conditions/diseases(for example,
were abstracted from the patient treatment charts. diabetes, hypertension, etc.). 88.9% of the patients had DDI with at least
The DDIs were identified with multiple(three) online drug interac­ one anticancer drug, excluding the interactions between the drugs used
tion checker softwares like “Drugs.com”, “Epocrates Interaction Check” to treat non-neoplastic conditions/diseases. The highest number of DDIs
software and “Medscape Drug Interactions Checker”. These 3 softwares involving antineoplastic medications in any given patient was 12. DDIs
were selected upon their evidence-based sensitivity, specificity, quality involving medications used for the treatment of non-cancerous diseases
of documentation of DDIs, comprehensiveness, and user-friendliness was observed in 83.3% of patients, out of which the highest number of
[13-15]. These softwares are freely available online. interactions identified in any given patient was 15.
“Drugs.com” stratifies DDIs as “Major (Highly clinically significant. Out of 473 interactions, 237(50.1%) DDIs were attributable to
Avoid combinations; the risk of the interaction outweighs the benefit)”, pharmacodynamic mechanisms of action. 126(26.6%) DDIs involved
“Moderate (Moderately clinically significant. Usually avoid combina­ pharmacokinetic mechanisms and 110(23.3%) involved unknown
tions; use it only under special circumstances)”, “Minor (Minimally mechanisms. “Drugs.com” software had unraveled 279 interactions, of
clinically significant. Minimize risk; assess risk and consider an alter­ which 20 were deemed as “Major” in terms of severity, 231 were iden­
native drug, take steps to circumvent the interaction risk and/or insti­ tified as “Moderate” in severity, and 28 were determined as “Mild” in
tute a monitoring plan)”, and “Unknown (No interaction information severity. “Epocrates Interaction Check” had identified 334 DDIs, out of
available)”. which 116 were deemed as “Contraindicated” category and 218 were
“Epocrates Interaction Check” stratifies DDIs into “Contraindicated”, determined to be of “Monitor/ Modify” category. “Medscape Drug In­
“Avoid/use alternative”, “Monitor/modify treatment”, “Therapeutic teractions Checker” identified 185 DDIs, of which 11 have been
advantage” and “Caution Advised”. DDIs between two anticancer drugs considered as “Serious/Use Alternative” category, 130 have been

2
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

Table 1 in the frequency of interactions were identified as cisplatin and pacli­

Demographic and clinical characteristics of the elderly cancer patients. taxel, which were responsible for 27.1% and 14.5% interactions,
Characteristics N ¼ 126n(%) respectively. The 10 most frequent DDIs involving medications used in
treating non-neoplastic conditions are shown in Table 4. 119 such DDI
Age
<70 years 99(78.57) pairs were observed.
>70 years 27(21.42) Twenty-eight QT-protracting DDIs were identified in 27 patients
Sex (21.4%) in our present study. The most frequent QT-protracting DDIs
Males 83(65.8) identified were cisplatin + pantoprazole (n = 9), ondansetron + tra­
Females 43(34.1)
Tumor types
madol (n = 9), ondansetron + oxaliplatin (n = 5), and doxorubicin +
(A)Solid tumors ondansetron (n = 4) (Table 5). Bivariate correlation (Pearson) test was
Lung 31(24.6) performed between the age, number of health problems, total number of
Oropharynx 17(13.4) medications, and the total number of interactions. The total number of
Breast 10(7.9)
interactions were positively correlated with the total number of medi­
Gall bladder 6(4.7)
Stomach 7(5.5) cations (r = 0.801, P < 0.001) and the number of health problems (r =
Cervix 5(3.9) 0247, P < 0.01) (Table 6). Multivariate logistic regression analysis
Ovary 5(3.9) showed that an increased number of medications was independently
Larynx 3(2.3) associated with the occurrence of DDIs involving anticancer agents (OR
Hypopharynx 3(2.3)
= 2.50; CI: 1.40–4.44; p < 0.01) (Table 6).
Urinary bladder 3(2.3)
Pancreas 3(2.3)
Prostate 3(2.3) Discussion
Sarcoma 2(1.5)
Others 4(3.1)
The present study has detected a high prevalence of DDIs in 97.6% of
(B)Hematological tumors
Leukemias 3(2.3) elderly patients afflicted with cancer taking antineoplastic chemo­
Lymphoma (Hodgkin’s & Non-Hodgkin’s) 9(7.14) therapy. Earlier studies solely involving older patients with cancer
Multiple myeloma 12(9.5) showed wide variation in the frequency of such DDIs with reporting
Comorbidities rates varying from 16% to 75% [7,8,16-18]. This variability may be
Diabetes 26(20.6)
attributed to different methodologies employed across studies, the
Hypertension 36(28.57)
Arthritis 18(14.2) diverse nature of the populations scrutinized, as well as the inclusion of
CAD 17(13.4) AzCERT (CredibleMeds) risk-stratification scheme, and differences in
Asthma 10(7.9) software sensitivity and specificity employed for scrutinizing the DDIs.
Depression 9(7.1)
The majority of the DDIs (48.7%) were deemed to be “Moderate” in
Epilepsy 9(7.1)
Acid peptic disorders 1(0.7) severity according to “Drugs.com”, 46.1% were considered to be
Neuropathy 8(6.3) “Monitor/Modify treatment” according to “Epocrates Interaction
Hepatitis B & C 4(3.17) Check”, and 6% were considered as “Serious-Use Alternative” as per
“Medscape Drug Interaction Checker”. Our findings are in contrast to a
study conducted by Faria et al., who reported that 50% of severe DDIs
considered as “Monitor Closely” category, and 44 have been identified
required clinical intervention [19]. The proportion of DDIs involving
as “Minor” category(breakup is shown in Table 2). The breakup of all
anticancer medications in our present study was 53%, which is higher
potential DDIs detected by the three online softwares is described in
when compared with a study by Faria et al. (26.3%) [19]. This high
Fig. 1.
prevalence is mainly attributable to polypharmacy and associated
Table 3 summarizes the 10 most frequent DDIs involving anticancer
comorbidities in elderly cancer patients.
medications along with their clinical effect, levels of severity, mecha­
nisms, and frequencies. We identified 100 DDI pairs involving anti­
DDIs involving paclitaxel ± dexamethasone
cancer medications. The anticancer agents with the highest involvement
Among the drug interactions involving anticancer medications in our
present study, DDIs were most frequently associated with cisplatin and
Table 2
paclitaxel. In their study, Hou et al., found that pretreatment with
Characteristics of Drug-Drug Interactions(DDIs).
dexamethasone reduced the inhibitory effect of paclitaxel in ovarian
Characteristics of DDIs n carcinoma xenografted mice [20]. It’s a theoretical interaction, there­
Interactions fore, clinicians must be alert to recognize the potential for interaction
Total no. of DDIs 473 and monitor for evidence of reduced therapeutic responses to paclitaxel
No. of DDIs involving anti-cancer drugs 247
when administered concomitantly.
No. of DDIs between adjuvant drugs 226
Significance
According to “Drugs.com”(n = 279) DDIs involving cisplatin ± dexamethasone
Considered as “Major” DDIs 20
Considered as “Moderate” DDIs 231 When interacting with dexamethasone, platinum coordination
Considered as “Mild” DDIs 28
According to “Epocrates Interaction Check”(n = 334)
compounds have similar outcomes, i.e., elevating the risk of infections
Considered as “Contraindicated or Avoid/Use Alternative” 116 and hypokalemia. There are established case reports regarding cisplatin-
Considered as “Monitor/Modify Treatment” 218 induced hypokalemia [21]. In addition to this, if dexamethasone is given
According to “Medscape Drug Interactions Checker”(n = 185) as a premedication, there is a proclivity towards increased risk of hy­
Considered as “Serious/Use Alternative” 11
pokalemia. Physicians should therefore monitor the potassium levels
Considered as “Monitor Closely” 130
Considered as “Minor” 44 regularly and educate the patients to report immediately if they develop
Mechanism of interaction symptoms like weakness, lethargy, or cramps. In the older population,
Pharmacokinetic 126 the risks of bone marrow suppression are higher than in other age
Pharmacodynamic 237 groups. Furthermore, the common adverse effects of such interactions
Unknown 110
include neutropenia, anemia, and thrombocytopenia [22].

3
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

Fig. 1. A Venn diagram showing the breakup of all

potential DDIs(473) detected from the three online
softwares engaged in this study. Area 1 = Potential
DDIs detected only by Drugs.com, Area 2 = Potential
DDIs Detected only by Epocrates Interaction Check,
Area 3 = Potential DDIs detected only by Medscape
Drug Interactions Checker, Area 4 = Potential DDIs
detected by both Drugs.com & Epocrates Interaction
Check, Area 5 = Potential DDIs detected by both
Epocrates Interaction Check & Medscape Drug In­
teractions Checker, Area 6 = Potential DDIs detected
by both Drugs.com & Medscape Drug Interactions
Checker, Area 7 = Potential DDIs detected by all three
softwares(Drugs.com, Epocrates Interaction Check &
Medscape Drug Interactions Checker).

Table 3
The 10 most common DDIs involving anti-neoplastic drugs.
Interaction Clinical effect Levels of severity Mechanism Frequency(N(%))

Dexamethasone + Paclitaxel Lowering of serum paclitaxel levels Moderatea Pharmacokinetic 18(14.2)

Cisplatin + Dexamethasone Increased risk of infection & hypokalemia Moderatea Unknown 16(12.6)
Carboplatin + Dexamethasone Increased risk of infection & hypokalemia Monitor/Modifyb Unknown 15(11.9)
Carboplatin + Filgrastim Decreased efficacy of filgrastim Moderatea Pharmacodynamic 8(6.34)
Cisplatin + Telmisartan Increased cisplatin levels Monitor/Modifyb Pharmacodynamic 8(6.34)
Cisplatin + Pantoprazole Increased risk of hypomagnesemia Moderatea Unknown 8(6.34)
Filgrastim + Paclitaxel Decreased efficacy of filgrastim Moderatea Pharmacodynamic 7(5.5)
Cyclophosphamide + Ondansetron Decreased effect of cyclophosphamide Minora Unknown 7(5.5)
Dexamethasone + Gemcitabine Increased risk of infection Caution advisedb Pharmacodynamic 6(4.7)
Cisplatin + Mannitol Increased cisplatin levels & risk of infection Avoid/Use Alternativeb Unknown 6(4.7)

DDIs = Drug-Drug Interactions.

a
Levels of severity as per “Drugs.com” software;.
b
Levels of severity as per “Epocrates Interaction Check”.

DDIs involving cyclophosphamide ± ondansetron shows the beneficial DDI between ondansetron and dexamethasone in
obviating nausea and vomiting 24 h following chemotherapy [25].
The interaction between cyclophosphamide and antiemetic ondan­ Moreover, it was evident that the combination of ondansetron and
setron has been established in a study [23]. Ondansetron altered the dexamethasone protects against delayed-onset nausea and vomiting in
systemic exposure of cyclophosphamide upon concomitant administra­ high-risk cancer patients.
tion. The clinical significance of this interaction is not known. Georgy
et al., have described in their study that the choice of an antiemetic
DDIs involving metformin ± dexamethasone
should be based on their safety, efficacy, and risk of DDIs [24].

The interaction between metformin and dexamethasone is deemed

DDIs involving ondansetron ± dexamethasone harmful in diabetic cancer patients. There is evidence that dexametha­
sone impairs insulin sensitivity and glucose tolerance [26,27]. Moni­
In our study, “Medscape Drug Interactions Checker” identified the toring of glycemic control is recommended following initiation or
interaction between ondansetron and dexamethasone. The clinical effect discontinuation of dexamethasone and the dosage of metformin needs to
would be decreased levels of ondansetron. In contrast to this, a study be adjusted, if necessary. However, this drug combination may be

4
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

Table 4
The 10 most common DDIs involving drugs for non-neoplastic diseases.
Interaction Clinical effect Levels of severity Mechanism Frequency(N(%))
c
Dexamethasone + Ondansetron Decreased levels of ondansetron Monitor Closely Pharmacokinetic 24(19)
Dexamethasone + Metformin Decreased effect of metformin Moderatea Pharmacodynamic 8(6.3)
Metformin + Ondansetron Increased metformin levels/lactic acidosis Caution advisedb Pharmacokinetic 7(5.5)
Metformin + Ranitidine Increased metformin levels/lactic acidosis Moderatea Pharmacokinetic 7(5.5)
Ondansetron + Tramadol Prolongation of QT interval Majora Pharmacodynamic 6(4.7)
Dexamethasone + Telmisartan Decreased effect of telmisartan Moderatea Pharmacodynamic 5(3.9)
Dexamethasone + Tramadol Decreased levels of tramadol/ seizures Caution advisedb Pharmacokinetic 5(3.9)
Glimepiride + Metformin Increased risk of hypoglycemia Moderatea Pharmacodynamic 5(3.9)
Insulin + Metformin Increased risk of hypoglycemia Moderatea Pharmacodynamic 5(3.9)
Aprepitant + Dexamethasone Increased levels of corticosteroid Moderatea Pharmacokinetic 4(3.1)

DDIs = Drug-Drug Interactions.

a
Levels of severity as per “Drugs.com” software;.
b
Levels of severity as per “Epocrates Interaction Check”.
c
Levels of severity as per “Medscape Drug Interactions Checker”.

Table 5
Top 5 QT-interval protracting drug-drug interactions(QT-DDIs) pairs.
a
QT-DDIs TdP Risk Therapeutic Class Frequency
Drug 1 Drug 2 Drug 1 Drug 2 N(%)b

Ondansetron-Tramadol KR PR Antiemetic Opioid 9(32)

Ondansetron-Oxaliplatin CR CR Antiemetic Antineoplastic 5(17.8)
Epirubicin-Ondansetron PR KR Antineoplastic Antiemetic 2(7.1)
Fluconazole-Ondansetron KR KR Antifungal Antiemetic 2(7.1)
Hydrochlorothiazide-Pantoprazole CR CR Diuretic PPI 2(7.1)

KR = Known Risk of TdP; PR = Possible Risk of TdP; CR = Conditional Risk of TdP;.

PPI = Proton Pump Inhibitor.
a
TdP risk based on AzCERT QT(CredibleMeds) Drugs Lists(1–3);.
b
Percentage calculated from total number of QT-DDIs = 28.

elderly because of life-threatening signs of respiratory distress, hyper­

Table 6
ventilation, and irregular heartbeat. In our study, combinations of
Univariate & Multivariate Analysis.
metformin with drugs like ondansetron and ranitidine raise the met­
Serial Characteristic(s) Unadjusted OR(p- Adjusted OR(p- formin levels predisposing to lactic acidosis [29]. Ondansetron and ra­
No value, 95% CI) value, 95% CI)
nitidine inhibit Organic Cation Transporter (OCT) and Multidrug and
1 Age 1.74 – Toxin Extruders (MATE), thereby elevating the risk of
(0.48, 0.36–8.31)
metformin-induced lactic acidosis [30,31]. Hence, if metformin and
2 Sex 0.9 –
(0.89, 0.29–2.95)
ranitidine must be used together, slow and cautious titration of met­
3 Number of health 0.59 0.80 formin dosage is recommended.
problems (0.136, 0.29–1.18) (0.626, 0.32–1.99)
4 Diabetes 0.23 0.14
(0.012, 0.07–0.73) (0.018, 0.03–0.71) QT-Protracting DDIs with anticancer agents
5 Hypertension 1.0 –
(1.00, 0.29–3.42)
Elderly patients with cancer have several risk factors for developing
6 CAD 0.52 –
(0.36, 0.13–2.11) arrhythmias. The prevalence of QT-protracting DDIs in our study was
7 Asthma 0.79 – 31.7% which is higher when compared to a study by Khan et al., where
(0.83, 0.09–39.01) they reported 21.7% [32]. QT-prolonging DDIs could lead to fatal con­
8 Arthritis 2.66 –
sequences like TdP. Polypharmacy is a major risk factor for DDIs arising
(0.36, 0.33–21.53)
9 APD 0.87 –
out of QT-extension [32]. Concerning QT-protracting DDIs, there is a
(0.89, 0.09–7.61) scarcity of information about the hazards of concomitant use of
10 Depression 0.74 – QT-extending medications, and how these DDIs should be managed.
(0.784, 0.08–6.60) DDIs leading to QT interval extension should be avoided in patients.
11 Number of drugs 1.93 2.50
In this present study, the total number of DDIs is positively correlated
(0.006, 1.21–3.09) (0.002, 1.40–4.44)
with the total number of medications and the number of health prob­
OR = Odds Ratio; CAD = Coronary Artery Disease; APD = Acid Peptic Disease. lems. This implies that as the number of drugs increases, there is an
elevated risk of DDIs in elderly cancer patients receiving antineoplastic
beneficial in patients suffering from multiple myeloma. Currently, chemotherapy. This finding is consistent with a few other studies
clinical trials are going on to assess the activity of metformin when given documented in the medical literature, in which an increase in the
with high doses of dexamethasone in refractory multiple myeloma. number of drugs administered have been associated with increments in
Metformin in combination with dexamethasone displayed antimyeloma DDIs [7,8,16]. Hersh et al., stated that risk factors for polypharmacy in
activity in-vitro and in-vivo xenograft models [28]. elderly patients with cancer are the number of health problems, inpa­
tient admission, and prescriptions to treat adverse drug reactions with
DDIs involving metformin ± ondansetron/ranitidine other medications [33]. The consequences of polypharmacy are adverse
drug reactions, DDIs, elevated risk of morbidity, and mortality.
Drug interactions that lead to lactic acidosis should be avoided in the It is likely that every physician and pharmacist cannot remember and

5
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

understand all potential DDIs and therefore cannot take corrective ac­ forestalling and identification of DDIs.
tions accordingly. They may be more familiar with the drugs used in Thirdly, elderly geriatric patients with cancer (or their family
their specialty but not with drugs used in other specialties. Frequent members and carers) have a role in educating themselves and creating
launches of new drugs and approval of new indications for marketed self-awareness about the hazards/warning signs and symptoms of DDIs
medicines make recognition of the occurrence of DDIs more difficult for (whether administered/self-administered).
health care professionals. To cope with that, several DDI screening Fourthly, clinicians have the privilege of employing the various on­
programs or databases have been developed and implemented as clinical line DDI Checking softwares to scrutinize the possible DDIs of prescribed
decision support tools. One of the tools that clinicians trust to review medications and institute necessary modifications accordingly. Physi­
patients’ medication sheets for DDIs is computerized DDI software. DDI cians must educate and alert the patients about the possible hazards of
checking/screening programs are widely used to identify potentially medication interactions and the potential risks of self-medication.
harmful drug interactions in the inpatient and outpatient setting. What Fifthly, there is a pressing need to harmonize and standardize the
is important is that these programs vary in accuracy and the information various online DDI Checking softwares and other DDI-related resources,
within interaction monographs. Previous studies have documented and make them more user-friendly, accessible, comprehensive, and
discrepancies(lack of consistency in the inclusion and grading of major algorithmic in approach (with inbuilt risk-stratification and alerting
drug interactions) across various DDI checking databases/programs systems).
[34-36]. Information pertaining to the interacting drugs and epidemi­ Finally, the importance of automated modalities like computerized
ological study resource availability and reliability is critical to each prescribing order entry as well as knowledge/non-knowledge-based
interaction monograph. Clinical relevance of DDIs evaluated with DDI clinical decision support tool workflows for assisting with decision-
software is a special concern because the software does not consider the making by clinicians (medication prescribers) cannot be over­
patient’s characteristics, dosing schedule, and precautions taken by the emphasized though their procurement costs could be staggering and
clinicians. Therefore, DDIs detected by the electronic databases may be hence, their availability is currently restricted to only a handful of
over-detected compared to the clinician’s assessment. Studies reported a centers [38].
high level of discrepancy between the number of DDIs identified by
electronic software and the number of clinically relevant DDIs assessed Conclusions
by a clinician. DDI checking databases/programs may invoke evidence
from a study without a control group for interaction to identify con­ Geriatric cancer patients are at high risk of DDIs ascribable to pol­
founding factors. Out of the 3 online DDI checking programs engaged by ypharmacy. So the treating oncologist must carefully review and cut
us for this study, Drugs.com included references to document DDI sci­ down on the list of medications before initiating antineoplastic
entific evidence. The disparity between electronic software databases in chemotherapy. The treating physician may utilize the online DDI
the severity rating of identified DDIs can be explained by the inconsis­ checking softwares to pinpoint the potential DDIs and can modify the
tency of evidence and different criteria for the classification of DDI medications judiciously. An integrative and algorithmic approach with
severity by various softwares [37]. A wise idea is to check more than one the inclusion of geriatricians, oncologists, cardiologists, general practi­
DDI checking databases/programs and compare their results to obtain tioners, and clinical pharmacologists/pharmacists is essential to opti­
optimum sensitivity and specificity. Besides, the judgment of the clini­ mize drug therapy in older patients afflicted with cancer.
cian is of utmost importance to discern relevant from irrelevant
interactions. Funding statement
The strengths of this study were that it was conducted prospectively
and the judgment of concerned clinicians in the context of potential This research did not receive any specific grant from funding
DDIs was deliberated upon and taken into account regarding their agencies in the public, commercial, or not-for-profit sectors.
clinical relevance to a certain extent. However, this study is primarily
limited by the small number of patients. In addition, our study findings
may not be applicable in other geographical settings as it was conducted Declaration of Competing Interest
in a single, large academic tertiary care teaching hospital setting.
Because convenience sampling was used, DDIs may be over or under- The authors declare that they have no known competing financial
represented for specific agents and indications. The clinical outcome interests or personal relationships that could have appeared to influence
was not assessed in these patients and it is not known if the detected the work reported in this paper.
potential DDIs caused adverse patient outcomes.
Acknowledgements
Proposals for attenuating the hazards of DDIs in geriatric oncology clinical
settings We thank Dr Sarika Palepu & Dr Arkapal Banerjee for their immense
help in statistical analysis of the data.
Regarding older patients with cancer, it is vital to effect steps to
curtail the risks arising out of DDIs, although in a few instances, such References
DDIs have been beneficial. Various measures have been advocated and
[1] Roberts D.C., McKay M.P., Shaffer A. Increasing rates of emergency department
clinicians have been advised to pay heed when prescribing medications visits for elderly patients in the United States, 1993 to 2003. Ann Emerg Med 2008;
that have the potential for inducing DDIs. 51(6): 769–774.
Firstly, the need for concomitant medication prescriptions must be [2] F. Salvi, V. Morichi, A. Grilli, R. Giorgi, G. De Tommaso, P Dessi-Fulgheri, The
elderly in the emergency department: a critical review of problems and solutions,
scrutinized thoroughly from all angles. In case, the use of concomitant Intern Emerg Med 2 (4) (2007) 292–301. Dec 1.
medications is imperative, biological and clinical monitoring may be [3] J.W. Park, J.L. Roh, S.W. Lee, S.B. Kim, S.H. Choi, S.Y. Nam, et al., Effect of
made mandatory. Many medications may need dosage modifications polypharmacy and potentially inappropriate medications on treatment and
posttreatment courses in elderly patients with head and neck cancer, J. Cancer Res.
and some agents may require discontinuation [14]. If the latter is not
Clin. Oncol. 142 (5) (2016) 1031–1040. May 1.
feasible, the offending medication may be substituted by another agent [4] E.D. Kantor, C.D. Rehm, J.S. Haas, A.T. Chan, E.L Giovannucci, Trends in
with less proclivity to induce DDIs. prescription drug use among adults in the United States from 1999 to 2012, JAMA
Secondly, the collaborative and educational role of clinical phar­ 314 (17) (2015) 1818–1830. Nov 3.
[5] B. Guthrie, B. Makubate, V. Hernandez-Santiago, T Dreischulte, The rising tide of
macologists/pharmacists is crucial in evidence-based scrutiny of the polypharmacy and drug-drug interactions: population database analysis
medications prescribed for each patient and hence, more efficient 1995–2010, BMC Med 13 (1) (2015) 74. Dec 1.

6
S.K. Ramasubbu et al. Cancer Treatment and Research Communications 26 (2021) 100277

[6] J.P. Turner, S. Shakib, N. Singhal, J. Hogan-Doran, R. Prowse, S. Johns, et al., [22] L. Balducci, H.J. Cohen, P.F. Engstrom, D.S. Ettinger, J. Halter, L.I. Gordon, et al.,
Prevalence and factors associated with polypharmacy in older people with cancer, Senior adult oncology clinical practice guidelines in oncology, JNCCN J National
Supportive Care in Cancer 22 (7) (2014) 1727–1734. Jul 1. Comprehensive Cancer Network 3 (4) (2005) 572–590. Jul 1.
[7] M.A. Popa, K.J. Wallace, A. Brunello, M. Extermann, L Balducci, Potential drug [23] C.J. Gilbert, W.P. Petros, J. Vredenburgh, A. Hussein, M. Ross, P. Rubin, et al.,
interactions and chemotoxicity in older patients with cancer receiving Pharmacokinetic interaction between ondansetron and cyclophosphamide during
chemotherapy, J Geriatr Oncol 5 (3) (2014) 307–314. Jul 1. high-dose chemotherapy for breast cancer, Cancer Chemother. Pharmacol. 42 (6)
[8] M.T. Puts, J. Monette, V. Girre, B. Costa-Lima, C. Wolfson, G. Batist, et al., Potential (1998) 497–503. Oct 1.
medication problems in older newly diagnosed cancer patients in Canada during [24] A. Georgy, J. Neceskas, S Goodin, Antiemetic care for patients with breast cancer:
cancer treatment, Drugs Aging 27 (7) (2010) 559–572. Jul 1. focus on drug interactions and safety concerns, American J Health-System
[9] M.L. Becker, M. Kallewaard, P.W. Caspers, L.E. Visser, H.G. Leufkens, B.H Stricker, Pharmacy 64 (21) (2007) 2227–2236. Nov 1.
Hospitalisations and emergency department visits due to drug–Drug interactions: a [25] Italian Group for Antiemetic Research, Dexamethasone alone or in combination
literature review, Pharmacoepidemiol Drug Saf 16 (6) (2007) 641–651. Jun. with ondansetron for the prevention of delayed nausea and vomiting induced by
[10] M. Pirmohamed, S. James, S. Meakin, C. Green, A.K. Scott, T.J. Walley, et al., chemotherapy, New England J Medicine 342 (21) (2000) 1554–1559. May 25.
Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 [26] B. Ludvik, M. Clodi, A. Kautzky-Willer, M. Capek, E. Hartter, G. Pacini, et al., Effect
820 patients, BMJ 329 (7456) (2004) 15–19. Jul 1. of dexamethasone on insulin sensitivity, islet amyloid polypeptide and insulin
[11] I. Etikan, S.A. Musa, R.S Alkassim, Comparison of convenience sampling and secretion in humans, Diabetologia 36 (1) (1993) 84–87. Jan 1.
purposive sampling, American J Theoretical and Applied Statistics 5 (1) (2016) [27] M.A. Greenstone, A.B Shaw, Alternate day corticosteroid causes alternate day
1–4. Jan 5. hyperglycaemia, Postgrad Med J 63 (743) (1987) 761–764. Sep 1.
[12] A.S. Acharya, A. Prakash, P. Saxena, A Nigam, Sampling: why and how of it ? [28] F.M. Zi, J.S. He, Y. Li, C. Wu, L. Yang, Y. Yang, et al., Metformin displays anti-
Indian J Medical Specialties 4 (2) (2013) 330–333. Jul 7. myeloma activity and synergistic effect with dexamethasone in in vitro and in vivo
[13] B. Suriyapakorn, P. Chairat, S. Boonyoprakarn, P. Rojanarattanangkul, xenograft models, Cancer Lett. 356 (2) (2015) 443–453. Jan 28.
W. Pisetcheep, N. Hunsakunachai, et al., Comparison of potential drug-drug [29] R.I. Misbin, The phantom of lactic acidosis due to metformin in patients with
interactions with metabolic syndrome medications detected by two databases, diabetes, Diabetes Care 27 (7) (2004) 1791–1793. Jul 1.
PLoS ONE 14 (11) (2019), e0225239. Nov 14. [30] Q. Li, H. Yang, D. Guo, T. Zhang, J.E. Polli, H. Zhou, et al., Effect of ondansetron on
[14] A. Mouzon, J. Kerger, L. D’Hondt, A Spinewine, Potential interactions with metformin pharmacokinetics and response in healthy subjects, Drug Metabolism
anticancer agents: a cross-sectional study, Chemotherapy 59 (2) (2013) 85–92. and Disposition 44 (4) (2016) 489–494. Apr 1.
[15] S.M. Janković, A.V. Pejčić, M.N. Milosavljević, V.D. Opančina, N.V. Pešić, T. [31] N.M. Maideen, A. Jumale, R Balasubramaniam, Drug interactions of metformin
T. Nedeljković, et al., Risk factors for potential drug-drug interactions in intensive involving drug transporter proteins, Adv Pharm Bull 7 (4) (2017) 501. Dec.
care unit patients, J Crit Care 43 (2018) 1–6. Feb 1. [32] Q. Khan, M. Ismail, S Khan, Frequency, characteristics and risk factors of QT
[16] V. Girre, H. Arkoub, M.T. Puts, C. Vantelon, F. Blanchard, J.P. Droz, et al., Potential interval prolonging drugs and drug-drug interactions in cancer patients: a
drug interactions in elderly cancer patients, Crit. Rev. Oncol. Hematol. 78 (3) multicenter study, BMC Pharmacology and Toxicology 18 (1) (2017) 75. Dec 1.
(2011) 220–226. Jun 1. [33] L.R. Hersh, K. Beldowski, E.R Hajjar, Polypharmacy in the geriatric oncology
[17] A. Alkan, A. Yaşar, E. Karcı, E.B. Köksoy, M. Ürün, Şenler FÇ, et al., Severe drug population, Curr Oncol Rep 19 (11) (2017) 73. Nov 1.
interactions and potentially inappropriate medication usage in elderly cancer [34] A.I. Vitry, Comparative assessment of four drug interaction compendia, Br J Clin
patients, Supportive Care in Cancer 25 (1) (2017) 229–236. Jan 1. Pharmacol 63 (6) (2007) 709–714.
[18] T.T. Yeoh, X.Y. Tay, P. Si, L Chew, Drug-related problems in elderly patients with [35] P. Vonbach, A. Dubied, J.H. Beer, S Krähenbühl, Recognition and management of
cancer receiving outpatient chemotherapy, J Geriatr Oncol 6 (4) (2015) 280–287. potential drug-drug interactions in patients on internal medicine wards, Eur J Clin
Jul 1. Pharmacol 63 (11) (2007) 1075–1083.
[19] C. de Oliveira Faria, C.M. Reis, A.G. Santos, A.M Reis, Interações Medicamentosas [36] J. Fish, Drug-drug interactions. A guide to identifying and managing important
na Farmacoterapia de Idosos com Câncer atendidos em um Ambulatório de Onco- drug interactions, J Pharm Soc Wisc 3 (2007) 16–25. Jul.
Hematologia, Revista Brasileira de Cancerologia 64 (1) (2018) 61–68. Mar 30. [37] T. Roblek, T. Vaupotic, A. Mrhar, M Lainscak, Drug-drug interaction software in
[20] W.J. Hou, J.H. Guan, Q. Dong, Y.H. Han, R Zhang, Dexamethasone inhibits the clinical practice: a systematic review, Eur J Clin Pharmacol 71 (2) (2015) 131–142.
effect of paclitaxel on human ovarian carcinoma xenografts in nude mice, Eur Rev Feb 1.
Med Pharmacol Sci 17 (21) (2013) 2902–2908. Nov. [38] R.T. Sutton, D. Pincock, D.C. Baumgart, D.C. Sadowski, R.N. Fedorak, K.I Kroeker,
[21] M. Mohammadianpanah, S. Omidvari, A. Mosalaei, N Ahmadloo, Cisplatin-induced An overview of clinical decision support systems: benefits, risks, and strategies for
hypokalemic paralysis, Clin Ther 26 (8) (2004) 1320–1323. Aug 1. success, NPJ Digital Medicine 3 (1) (2020) 1–10. Feb 6.