Persons 65 years of age and older are the fastest growing segment of the US population. In 1995 this group comprised 12.8% (33.6 million) of the population and by the year 2030 the estimates are that this number will reach 20.1% (70.2 million). The over 75-year-old group will triple in size by 2030. The number of people aged 85 years or older will double in the same period.[1] The average person 65 years of age can expect to live another 15 years and remain functionally independent for most of that time. The 75- and 85-year-old patients have an average life expectancy of 10 and 6 years, respectively, and will often function independently.[2–6] 50% of all cancers occur in those aged 65 years or older.[7] The risk of developing cancer increases with the passage of each decade of life. Therefore older persons bear the brunt of the cancer burden.[1,5]

It has been demonstrated that the pattern of care and treatment decisions change significantly with increasing patient age.[8,9] Physicians may undertreat elderly patients with chemotherapy or may be less likely to refer elderly patients for treatment, often because of the fear of toxicity.[10,11] Studies that have addressed chemotherapy toxicity with age have shown that the elderly often tolerate the regimens as well as younger patients.[12–20] As the elderly population rapidly increases there will be an increasing need for investigation, physician education and therapies for this group.[21,22]

There has been an increasing trend over the past 20 years toward the use of oral chemotherapy in the treatment of patients with a variety of malignancies.[23] Cost has been a significant impetus for this development. Oral therapy can eliminate charges for items such as administration and nursing, central intravenous catheter and infusion pump costs, and costs related to adverse effects of intravenous administration such as line sepsis.[24] Patient preferences and quality of life issues are also important factors for the development of oral chemotherapy. More than 90% of patients prefer an oral agent, provided efficacy is maintained. The reasons include convenience, concerns or difficulties with intravenous access lines, or ability to control the chemotherapy administration environment.[25] This article will review pertinent aspects of pharmacology of antineoplastic agents with emphasis on oral chemotherapy in elderly patients.

Factors that may affect the pharmacokinetics of chemotherapeutic agents in the elderly in comparison with younger populations include: a greater number of disease states, more medications, more drug interactions, and greater variability in nutritional status and in underlying chronic health status.[26,27] Prolonged survival of patients poses an interesting question: can pharmacokinetic data from patients in their sixties be applied to patients in their seventies or eighties? It is most likely, that individuals experience age-related changes throughout their lifetime that may be difficult to quantify. The reviews of the pharmacokinetic data of chemotherapy agents in the elderly generally have multiple limitations. Most data concerning pharmacokinetics in older patients is limited to the inclusion of older patients in studies that have a wide range of patient ages. The majority of available data are therefore from studies which are not specifically targeted at an older age group.

For oral formulations to be effective they must have adequate bioavailability. Bioavailability is the rate and extent to which a drug is absorbed into the systemic circulation.[24] In addition, the limitations of saturability and structural stability in gastric and intestinal pH are very important. Saturable absorption can be demonstrated by oral etoposide, with a 76% bioavailability of etoposide 100mg, while a 400mg dose was only 48% bioavailable.[28]

1. Compliance

Patient non-compliance is potentially a major obstacle for orally formulated chemotherapy. There are a number of studies evaluating compliance with regard to oral chemotherapy regimens.[29–32] The degree of compliance is highly variable and can be affected by the complexity of a regimen. This needs to be taken into account when designing treatment programmes. There are data which indicate that compliance with oral chemotherapy may influence survival.[33] It has been shown that incomplete treatment in the adjuvant treatment of patients with breast cancer results in a markedly inferior disease-free survival.[34] Factors associated with higher rates of non-compliance include lower socio-economic status or treatment in a community-based setting as well as dosage per day.[35] Pharmacokinetic analysis showed actual compliance was less than half that suggested by patient self-report. Measures designed to increase compliance, including patient education, home psychological support, and exercises in pill taking, increased compliance nearly 3-fold.[36,37]

To overcome problems of compliance in the elderly, providers should prescribe a simple dosage regimen for all medications (preferably 1 or 2 doses daily), help the patient select cues that assist them in remembering to take doses (time of day, meal-time, or other daily rituals), provide devices to simplify remembering doses (medication boxes), and regularly monitor compliance. A variety of compliance aids are available to help patients organise their medications (e.g. plastic boxes) or remember dose times (alarms). Medication packaged in standard pharmacy bottles should be identified with special labels, or dose charts can be provided to check the daily schedule. Single-unit doses, widely used in hospitals, may be cumbersome for elderly patients who have difficulty opening the foil-backed wrappers. Medication boxes with compartments that are filled weekly by the patient, family member or a home healthcare provider are useful organisers that simplify the patient’s responsibilities for self-administration. Microelectronic devices such as Medication Event Monitoring systems (MEMS) can provide feedback that shows patients whether they have been taking doses as scheduled and are also useful in study situations.[37]

2. Clinical Pharmacology

Elderly patients are the largest users of pharmaceuticals.[38,39] However, most studies are conducted in patients younger than 55 years.[40] This can make decision making, particularly in regards to dosage, quite difficult. This, in addition to age-related changes in pharmacokinetics and pharmacodynamics, may contribute to the increased incidence of drug toxicity with age. There are a number of physiological changes which accompany human aging. These include an increase in body fat, decrease in lean body mass and decrease in total body water.[41–44] Elderly patients with cancer have a number of significant comorbid illnesses which also may effect the disposition and efficacy of the drug.[45,46]

3. Pharmacokinetics/Pharmacodynamics

Pharmacokinetics are important to the clinical use of chemotherapy based on two principles.[47,48] Firstly, the optimal clinical response can be obtained if drug concentrations are maintained within a given range of concentrations. Secondly, genetic, environmental, pathophysiological, or drug-drug interactions may result in a remarkable difference among patients with regard to elimination of a drug from the body. Consequently, there may be a wide difference among patients in the dose required to achieve an optimal clinical response.

Pharmacodynamics is the study of the interactions of drugs with cellular and molecular targets. Factors that may be altered with age include transmembrane transport, and sensitivity to drug action. The items of particular concern include multidrug resistance from accumulation of P-glycoprotein in cellular membranes, cellular anoxia, increased DNA repair in older cells and the synthesis of abnormal proteins.[49] Pharmacodynamics is also the study of dose-response relationships. The endpoints that we are concerned about are toxicity and response. For phase-specific agents, the cells must be in the correct phase of the cell cycle. Therefore, response may be improved by increasing drug exposure. Usually in drugs in which phase is not important, the drug effect is more related to the duration of exposure above a threshold concentration. However, schedule dependence can still be a concern.[50,51]

In regards to myelotoxicity, hematopoietic recovery in older patients may be compromised by exhaustion of the pluripotent stem cell reserves. Hematopoietic stress causes a rapid decline in the concentration of stem cells in older but not younger mice.[52–54] In healthy volunteers, the number of peripheral blood progenitor cells was 2-fold higher in individuals aged between 20 and 30 years as compared with individuals aged between 70 to 80 years.[55] In an analysis of the experience of the Eastern Cooperative Oncology Group, myelotoxicity was not enhanced in patients older than 70 years with a good performance status. However, this was a highly selected group in a retrospective analysis.[17,54]

In a series of phase I studies no significant correlation between drug clearance and age was noted for individual drugs or the group of patients as a whole.[56] There was also no significant difference between the older (>65 years) and younger age groups with regard to dose or toxicity.[56] A retrospective study of the European Organisation for Research and Treatment of Cancer phase II trials showed that elderly patients did not show greater toxicity than younger patients.[57] Similar results were obtained from an analysis of patients from the Illinois Cancer Center (USA).[13]

3.1 Absorption

There are a number of age-related changes in the gastrointestinal tract which can affect drug absorption, including reduced gastric acid secretion, reduced gastric emptying time, reduced gastrointestinal motility and reduced splanchnic blood flow. Additionally, loss of absorption surface area may be a result of prior surgery.[40,58] Obviously, any such changes can impact on absorption and may result in altered toxicity and effectiveness of orally available chemotherapy drugs. Significant interpatient and intrapatient variability in bioavailability may be present as is exhibited with 6-mercaptopurine.[59,60] Finally, polypharmacy with multiple concomitant medications can alter absorption, through changes in binding, adsorption, pH and competition for sites.[61]

3.2 Distribution

Drug distribution is influenced by alterations in plasma proteins.[48] There are known age-associated decreases of 15 to 20% in plasma albumin. For highly protein-bound drugs this would result in a larger volume of distribution. Fat content doubles in the elderly from 15 to 30% of bodyweight, and intracellular water decreases from 42 % in the average 25-year-old to 33% in the average 75-year-old. This leads to an increase in the volume of distribution of lipid soluble drugs and a decrease in the volume of distribution of more polar drugs that primarily distribute to body water. This can lead to a lower peak concentration and prolonged terminal half-life.[58]

3.3 Metabolism

There is a lack of agreement regarding age-related changes in hepatic drug metabolising capability. However, there is consensus that liver size decreases with age.[48] Liver blood flow is reduced at a rate of 0.3 to 1.5% per year after the age of 25. This may lead to lower clearances of drugs which are highly dependent on blood flow for elimination.

Phase I metabolism occurs primarily via the cytochrome P450 (CYP) microsomal system, which consists of a number of isoenzymes. Phase II reactions are primarily conjugation reactions.

3.3.1 Cytochrome P450

These heme-based enzymes are grouped into various classes based on various types of metabolic processes and genetic homology. In addition to the liver, these enzymes are located in the small bowel, kidneys, lungs, and to a much lesser extent the brain. Genetic variability accounts for differing levels of enzyme activity through various pathways that may lead to clinically important pharmacodynamic differences among individuals.[27,62,63]

CYP enzymes known to be involved in drug metabolism are included in families I to IV. The major enzyme pathways of drug metabolism are the CYP3A4, CYP2D6, CYP1A2, and the CYP2C subfamily. See table I for oral chemotherapeutic agents that are substrates of CYP3A4 and P-glycoprotein. Many non-oncology drugs may compete and/or interfere with these enzyme systems and therefore may affect the metabolism of chemotherapeutic agents.[64] For a drug to be susceptible to pharmacokinetic drug interactions from enzyme inhibition or induction it generally must have at least 30% of its metabolism through that one enzyme substrate.[65] The potential for drug interactions is relatively high, particularly with the CYP3A4 enzyme. This enzyme is inhibited by a variety of commonly prescribed medications, and is involved in the metabolism of a variety of anticancer agents. Cyclophosphamide, ifosfamide, paclitaxel, docetaxel, etoposide, teniposide, vincristine, vinblastine, busulfan and tamoxifen are all substrates of CYP3A4 and may be significantly affected by common inhibitors of this enzyme. Variations in CYP3A4 activity have been demonstrated in different ethnic groups. There is a high frequency of this CYP3A4 variant in African Americans compared with US Caucasians and Taiwanese.[66]

Table I
figure Tab1

Oral chemotherapeutic agents that are substrates of CYP3A4 and P-glycoprotein[24]

Full size image

Genetic variation in the degree of activity expressed by various CYP pathways has been identified for some of the CYP enzymes. For example, variation in the enzyme activity expressed in a particular individual has been identified in humans for the CYP2D6 pathway. These variations are the result of genetic variation alone. Approximately 3% of the population are poor metabolisers through the CYP2D6 pathway, which is apparently inherited in an autosomal recessive pattern. Various ethnic populations have been studied with 2 to 10% of the populations estimated as poor metabolisers by CYP2D6. Populations of Japanese, Panamanian and Chinese people have extremely low rates of poor metabolisers through this pathway.[27,62,63,67] The risk of drug interactions for poor metabolisers may be a serious concern. Many β-blockers, antiarrhythmics, antidepressants and antihypertensives are substrates for this enzyme, which can lead to excessive pharmacological effects from standard doses.[27,65,67]

Age-related declines in these systems have been demonstrated in animal studies and some human trials. It has been demonstrated that CYP 1A2 shows a decrease of 20 to 25% in drug clearance in healthy elderly men and women compared with younger individuals.[68,69] This may result in decreased first-pass metabolism of drugs which are highly extracted by the liver, such as morphine. Other variables to be considered are the effect of age and diet, genetic polymorphisms and stereoselective drug metabolism.[69] Many alterations of the enzyme activity involve either drug-induced increases or decreases in metabolism.

Phase II reactions appear unaffected by age.[70] The biliary excretion of drugs has been studied, but no age-related alterations have been noted.[70]

3.4 Excretion

There are also age-related changes in excretory function. There is a gradual loss in renal mass and a decline in function with age. The combined weight of the kidneys declines 30% by age 90. This loss is primarily because of loss of cortical mass with relative preservation of the renal medulla. Glomerular sclerosis produces loss of capacity to perform ultrafiltration of plasma, which leads to a decrease in the glomerular filtration (GFR) rate by approximately 1 ml/min for every year over 40 years of age.[71–73] The importance of this decline was first emphasised in a study that detrmined that doses based on renal function lead to a more favourable therapeutic index.[12] The reduction in GFR is not reflected by an increase in serum creatinine levels because of the simultaneous loss of muscle mass which occurs with age. Estimations of creatinine clearance in the elderly may lead to errors in drug administration.[74]

Traditionally the serum creatinine level has been used to estimate the glomerular filtration function of the kidney, since creatinine is primarily filtered through the glomerulus; however, a small portion undergoes tubular secretion. In order to facilitate the estimation of glomerular clearance, various equations have been evaluated to calculate creatinine clearance based on the serum creatinine level and other factors. The two most common equations clinically utilised are the Cockcroft-Gault and Jelliffe equations.[75,76] The equations are less accurate in populations such as those with severe renal failure, patients with decreased muscle mass and the elderly. Many individuals lose muscle mass with age. In many elderly individuals, a low serum creatinine level of less than 1 mg/dl may actually reflect diminished muscle mass and diminished production of creatinine rather than exceptional renal function.

The decline in GFR with age translates into pharmacokinetic alterations of drugs which are excreted by the kidneys. Because of the physiological decline in renal function with age, chemotherapy agents which are primarily renally excreted must be used with extreme care in the elderly; standard doses may be too toxic. This is particularly true of the frail elderly. Table II lists suggested dosage modifications according to renal function and also shows fractional renal excretion for selected drugs.[54,77]

Table II
figure Tab2

Recommended dosage adjustments according to renal dysfunctiona and fractional renal clearance of selected chemotherapeutic drugs[54,77]

Full size image

4. Polypharmacy

Polypharmacy is quite frequent in the elderly population. Older ambulatory patients use 3-fold more medications than younger patients.[69] At least 90% of older patients use at least one medication, and the average is at least four medications per patient. Self medication with herbal remedies and other alternative therapies is becoming increasingly common. This large number of drugs also leads to a number of inappropriate medications and increased toxicity,[61] and can potentially lead to significant drug interactions, particularly those involving the CYP system.[65,78]

5. Specific Drugs

The drugs chosen for review represent a combination of newer chemotherapeutic agents (i.e. temozolomide) and those which have been used more commonly (i.e. etoposide). In addition, other medications are discussed that in the authors’ opinion may be commonly used in the future (i.e. topotecan). Tamoxifen is discussed as it is the most common oral hormonal agent used. Reviews are available that discuss the other available hormonal agents for breast cancer.[79,80]

5.1 Temozolomide

Temozolomide is an oral alkylating agent which is a second-generation imidazotetrazine and is the methyl analogue of mitozolomide. It is currently approved for the treatment of patients with malignant glioma and malignant melanoma.[81–84] Temozolomide has a starting dose of 200 mg/m2/day for 5 days every 28 days.[81]

The drug is rapidly and completely absorbed following oral administration. Peak plasma concentrations occur within 1 hour. Food reduces the rate and extent of temozolomide absorption. When taken with food the time to reach the peak plasma drug concentration (tmax) is delayed to 2.25 hours. A high-fat meal decreases the peak plasma drug concentration (Cmax) by 32% and the area under the concentration-time curve (AUC) by 9%. The potential for drug-protein binding interactions is relatively low because of minimal plasma protein binding of approximately 15%.

The pharmacokinetics of temozolomide follow a one-compartment model with first order absorption and elimination.[85] Temozolomide does appear to distribute widely to tissues including the cerebrospinal fluid.[86] Temozolomide clearance is increased with increased body surface area (BSA) for both genders. The population mean clearance for patients with glioblastoma or anaplastic astrocytoma was 11.2 L/h for males with BSA equal to 2.0m2and 8.8 L/h for females with BSA equal to 1.7m2. The interindividual variability in clearance was 15%, and the residual variability was 26%. Other factors investigated were age, gender, height, bodyweight, serum creatinine level, estimated creatinine clearance, serum chemistry data as indices of hepatic function and disease, smoking status and selected concomitant medications.[85]

The pharmacokinetics of temozolomide in patients with mild to moderate hepatic dysfunction are similar to those of patients with normal hepatic function. Patients with severe hepatic dysfunction have not been studied.

Although studies have not indicated significant differences in temozolomide pharmacokinetics, the data do indicate that patients over the age of 70 years did experience an increased incidence of grade 4 neutropenia and grade 4 thrombocytopenia versus patients younger than 70 years.[87] A summary of the pharmacokinetics of temozolomide is presented in table III.

Table III
figure Tab3

Pharmacokinetic parameters of oral temozolomide

Full size image

5.2 Thalidomide

Thalidomide was developed as a sedative in the late 1950s but was quickly removed from the market when its teratogenic effects were discovered. D’Amato et al.[91] demonstrated that thalidomide has potent antiangiogenic activity in vivo, a property that may be related to its mechanism of teratogenesis. In light of its excellent oral bioavailability and minimal adverse effects, thalidomide is a promising antiangiogenic agent for long term therapy in patients with vascular tumours.[91] The mechanism of action of thalidomide is not completely understood. In vitro and in vivo studies show that thalidomide selectively reduces levels of tumour necrosis factor alpha (TNF-α) by accelerating the degradation of TNF-α mRNA encoding protein. Recently, thalidomide has been recognised as an inhibitor of angiogenesis because of inhibition of basic fibroblast growth factor (bFGF). bFGF has been shown to stimulate limb growth and its inhibition may be the basis for the limb defects associated with thalidomide.[91]

Thalidomide is administered orally and is slowly absorbed from the gastrointestinal tract. The exact bioavailability has not been determined because of poor aqueous solubility of thalidomide. The AUC is proportional to the dose but at doses >200mg a flattening of the peak concentration curve is noted with an associated delay in the tmax. Following 200 and 400mg doses of thalidomide in healthy volunteers Cmax was 1.76 and 2.82 µg/ml, tmax was 3.5 and 4.3 hours, and AUC was 18.9 and 36.4 µg • h/ml, respectively. Administration of thalidomide with a high fat meal causes minor changes in AUC and Cmax but increases tmax to about 6 hours in healthy volunteers.[70,87] The exact mechanism of metabolism and elimination of thalidomide is not clearly understood. Thalidomide appears to undergo non-enzymatic hydrolysis in the plasma and is not hepatically metabolised. The half-life seems to be similar in all groups studied with an average half-life of 6 to 7 hours. Renal excretion plays a minor role: less than 0.7% of the dose is excreted in the urine as unchanged drug and thalidomide is undetectable in the urine 48 hours after a single dose.[92] In addition, only a small amount of thalidomide metabolites may be detected in the urine 12 to 24 hours after administration.[70] The pharmacokinetics of thalidomide in patients with renal or hepatic insufficiency have not been determined. Thalidomide has shown activity in patients with multiple myeloma and malignant gliomas.[93–95]

5.3 Idarubicin

Idarubicin is available in both intravenous and oral forms. It may have lower cardiotoxicity than doxorubicin and is well tolerated by the elderly.[96,97]

The oral formulation has 30% bioavailability.[98] The oral formulation undergoes first-pass hepatic metabolism. The active parent compound is metabolised to an active metabolite idarubicinol. The half-life of oral idarubicin is 5 to 24 hours and 13 to 60 hours for idarubicinol.

The bioavailability is independent of age and the pharmacokinetics are not altered in the elderly. Idarubicin pharmacokinetics were not effected by the presence of liver metastases but were related to the integrity of kidney function. Total body clearance is significantly decreased in patients with renal dysfunction.[99] There is a significant correlation between idarubicin plasma clearance and creatinine clearance. It was also found that idarubicin plasma clearance was lower in patients whose creatinine clearance was less than 60 ml/min [83.4 ± 18.3 L/h versus 122.8 ± 44.0 L/h in patients with normal renal function (p = 0.037)]. The idarubicinol terminal elimination half-life was significantly increased in patients with impaired kidney function [t½γ: 41.3 ± 10.1h in patients with normal renal function versus 55.8 ± 8.2h in patients with creatinine clearance less than 60 ml/min (p = 0.025)].

In patients with renal impairment dose adjustments may be considered. One recommendation is to give 75% of the standard dosage of idarubicin if the serum creatinine level is 2 mg/dl or greater. Similar to other anthracyclines, idarubicin dosage should be adjusted if patients have elevated bilirubin levels. Recommended dosage adjustments include administering 50% of the standard dosage if the total bilirubin level is 1.5 to 5.0 mg/dl or aspartate amino transferase (AST) level is 60 to 180 U/L. In addition, if the total bilirubin is greater than 5.0 mg/dl the dose should be withheld.

5.4 Topotecan

The oral formulation of topotecan is undergoing clinical trials. Topotecan is a topoisomerase I inhibitor which is approved for the treatment of patients with recurrent or refractory ovarian cancer. It has activity in small cell lung cancer and has some promising anti-leukaemia effects.[100,101]

Topotecan has a half-life of 3 hours with renal clearance accounting for 30% of the drug elimination as well as substantial biliary concentration. The standard intravenous dosage is 1.5 mg/m2/day for 5 days every 3 weeks. Topotecan dose modifications are not required for patients with hepatic dysfunction and normal renal function.[102] Dosage adjustments are required in patients with moderate to severe, but not mild, renal impairment (table IV). Guidelines for patients with creatinine clearance less than 20 ml/min are not established, and strong consideration should be given to withholding the drug in these patients.[103] Life-threatening myelosuppression can occur if this is not taken into account.

Table IV
figure Tab4

Dose recommendations (mg/m2/day) for intravenous topotecan for patients with renal impairment[103]

Full size image

The oral formulation is currently being developed. Oral topotecan is at least as effective as intravenous topotecan in the treatment of patients with relapsed, sensitive small cell lung cancer.[104] It has been investigated in phase I studies in dosage schedules of daily administration for 5 or 21 days. The dose-limiting toxicity is schedule dependent with myelosuppression in the shorter schedules and diarrhoea in the more prolonged treatments.

Topotecan demonstrates suitable bioavailability for oral treatment, with the bioavailability being 40%.[105] The pharmacokinetics of topotecan after oral administration are not significantly different than those after intravenous administration. Co-administration of the topotecan gelatin capsules with a high-fat breakfast leads to a small decrease in absorption rate but does not affect the extent of absorption.[105] The ratio of the AUC∞ during fasting to that during a high-fat meal was 0.93 ± 0.23 [90% confidence interval (CI) 0.83−1.03]. Cmaxof topotecan was similar after ingestion of the capsules with (10.6 ± 4.4 µg/L) or without food [9.2 ± 4.1 µg/L (p = 0.130)]. The tmax was significantly prolonged after food intake (median 3.1h, range 2.8−6.1) compared with fasted conditions (2.0h, range 1.1 to 8.1) [p = 0.013]. The absolute bioavailability of topotecan averaged 42 ± 13% (90% CI 37 to 47%). The apparent terminal half-life was significantly longer after administration of oral topotecan (3.9 ± 1.0h) than after intravenous administration [2.7 ± 0.4h (p < 0.001)].[105]

An interaction with probenecid has been demonstrated. By inhibiting renal tubular secretion, probenecid decreased renal and systemic clearance which led to an increase in topotecan systemic exposure.[106] Topotecan elimination through renal tubular secretion may have clinical relevance for the use of topotecan in patients with altered renal function.[106] There was a slightly, but not significantly, faster mean rate of absorption for total topotecan in the presence of ranitidine. The increase in Cmax was consistent with a slight increase in the rate of absorption.[107]

5.5 Fluoropyrimidines

Fluorouracil (5-fluorouracil; 5-FU) is an antimetabolite which has continued to be useful in the treatment of patients with solid tumours such as gastrointestinal adenocarcinoma, breast cancer, and squamous cell cancer of the head and neck. Fluorouracil has been administered by intravenous bolus and continuous infusions, and also oral administration. There is no evidence that there is an improvement in survival with any one schedule. Treatment with fluorouracil is characterised by considerable interpatient variability in pharmacokinetics, toxicity and responses. Fluorouracil has been combined with modulators such as folinic acid (leucovorin), methotrexate, cisplatin, thymidine, interferon and allopurinol.[108]

Absorption of fluorouracil in the upper gastrointestinal tract is unpredictable and erratic. A wide range of bioavailability values have been determined. Bioavailability of fluorouracil increases with increasing doses, indicating a saturable first-pass metabolism. It is transported actively across intestinal membranes at low concentrations and passive processes predominate at higher concentrations. The unpredictable and significant interpatient variability in absorption after oral administration may be partially because of differences in fluorouracil catabolism by dihydropyrimidine dehydrogenase (DPD) enzyme in the intestines and liver.

The significant interpatient variations in fluorouracil clearance, tumour response, and host toxicity may be explained in part by genetic differences in the activity of DPD enzyme. Age does not appear to be an influencing factor on the clearance of fluorouracil. There appears to be gender differences in disposition, because women have been shown to clear fluorouracil at significantly lower rates than men. DPD activation is not correlated with age or race but is influenced by gender with 15% lower activity in women.[109]

There have been a number of pharmacological approaches to reduce variability in the pharmacokinetics of oral fluorouracil. These are listed in table V.

Table V
figure Tab5

Fluoropyrimidines[108]

Full size image

The role of the kidney in the elimination of fluorouracil is usually negligible. However, when the DPD inhibitor eniluracil is used in combination with fluorouracil the contribution of renal function to excretion becomes more significant. An average of 77% of fluorouracil was excreted unchanged in urine in patients receiving the combination of fluorouracil and eniluracil.[110] This may make the elderly with decreased renal function more susceptible to fluoropyrimidine toxicities.

Capecitabine is a precursor of 5-deoxy-5-fluorocytidine and is metabolised by carboxylesterase, cytidine deaminase and intratumoral thymidine phosphorylase to fluorouracil. It has activity in breast and colorectal cancer and is associated with palmar-plantar erythrodysesthesia as seen with a continuous fluorouracil infusion.[111] The hand-foot syndrome occurs in approximately 40% of patients receiving capecitabine.[111] In a breast cancer trial of postmenopausal women it had minimal toxicity.[112]

Capecitabine pharmacology is not significantly affected by age, gender, BSA or creatinine clearance.[113] There is no significant effect on the pharmacokinetics of the drug in patients with hepatic dysfunction.[114] Food has a profound effect on the AUC and Cmax of capecitabine. The recommendation is that the drug should be administered with food as this was done in the clinical trials.[115]

Reigner et al.[115,116] evaluated the effect of Maalox® (aluminium hydroxide and magnesium hydroxide) on the pharmacokinetics of capecitabine. It appears that Maalox® administration may mildly affect the concentrations of capecitabine and metabolites; however, the differences are minor with no clinically significant difference expected with co-administration of these agents.[115,116]

5.6 Etoposide

Etoposide is a semisynthetic podophyllotoxin derivative, which is a topoisomerase II inhibitor used most often in geriatric patients. It is very useful in elderly patients with refractory non-Hodgkin’s lymphoma, lung cancer and ovarian cancer.[117–121] Traditionally, the agent is given through the intravenous route because of difficulties with absorption and tolerance of the oral product. Oral administration of etoposide has increased in recent years because of investigation of schedule-dependent differences in efficacy with prolonged oral administration versus intravenous doses over single or multiple days.[122]

Etoposide is available orally in a soft 50mg gelatin capsule formulated with citric acid, glycerin, purified water, and polyethylene glycol 400. Etoposide shows either biphasic or triphasic pharmacokinetic characteristics with an initial half-life (range) of 0.6 to 2 (0.25 to 2.5) hours, and a terminal half-life of 5.3 to 10.8 (2.9 to 19) hours. Oral etoposide absorption is typically erratic but approximately 50% (25 to 75%) of the dose is absorbed. Some data do suggest that doses greater than 200 to 400mg may have decreased absolute bioavailability, and that intrapatient variation in bioavailability can occur with repeat administration. The typical tmax reported for etoposide is 1 to 1.5 (0.75 to 4) hours. The reported volume of distribution of etoposide following intravenous administration is 7 to 17 L/m2.[77,123]

Patients with impaired renal function have decreased drug clearance rates of etoposide, therefore dosage should be reduced in proportion to the reduction in creatinine clearance. Oral etoposide bioavailability has shown substantial inter- and intrapatient variability and is not affected by food or concurrent intravenous chemotherapy. Myelosuppression and mucositis are the predominant toxicities. The pharmacology of oral etoposide shows that increasing etoposide concentration correlates with advancing age. Care must be taken in using this medication in older patients with a poor performance status. These patients are at higher risk for grade 4 toxicities.[119]

The issue of dosage adjustment for oral etoposide administration in patients with hepatic dysfunction is controversial. Etoposide is eliminated to some degree through hepatic metabolism and the CYP system, but guidelines for dosage adjustments based on liver dysfunction are difficult to establish based on the literature. Aita et al.[124] evaluated oral etoposide pharmacokinetics in 17 patients with hepatocellular carcinoma with a mean age of 65 (range 52 to 83) years. Patients were stratified into a group of ten with bilirubin level < 1.2 mg/dl (group 1), versus seven with bilirubin levels > 1.2 mg/dl (group 2). Group 2 had significantly higher mean bilirubin [2.8 vs 0.9 mg/dl (p = 0.0006)], higher mean AST [150 vs 47 IU/L (p = 0.02)], higher prothrombin time index [1.25 vs 1.05 (p = 0.02)], but no difference in albumin level when compared with group 1 at baseline. Patients received either etoposide 100mg orally or 50mg intravenously on day 1 or day 8 to compare intravenous and oral pharmacokinetics, followed by subsequent cycles of daily etoposide 100mg orally for 14 days every 3 weeks. The study identified a relatively high bioavailability in patients at a mean of 61%, with a wide range observed (17 to 95%), a mean volume of distribution of 8.41 L/m2 and mean half-life of 5.1 hours. In addition, the mean AUC observed was 27.0 µg • h/ml and mean protein binding was 93.2%. One parameter that did appear to be increased in group 2 relative to group 1 was the mean clearance of 1.1 L/h • m2 (not significant). When the groups were evaluated separately the pharmacokinetic parameters did not differ significantly, and were consistent between patients with or without hepatic dysfunction. In addition, no differences in haematological toxicity were noted based on bilirubin baseline values. This study shows that the pharmacokinetics of oral etoposide in patients with liver dysfunction in a primarily geriatric population do not significantly differ from other populations.[124]

5.7 Hydroxycarbamide

Hydroxycarbamide (hydroxyurea) is an urea derivative capable of inhibiting the incorporation of thymidine into DNA; however, the exact antineoplastic mechanism of action of hydroxycarbamide is not certain. Hydroxycarbamide has documented clinically significant activity in patients with chronic myelogenous leukaemia.[125]

Hydroxycarbamide is considered easily and readily absorbed after oral administration, with a bioavailability of approximately 100%. Peak concentrations are typically reached within 1 to 2 hours of administration. The elimination half-life appears to be 3.5 to 4.5 hours.[77,123,126]

A study by Rodriguez et al.[127] was designed to characterise the pharmacokinetics of oral and intravenous hydroxycarbamide. The mean age of the patients was 64 years (range of 28–88 years) with a variety of solid tumours. The mean oral bioavailability in 22 patients who fasted for 8 hours prior to administration was 108% with a coefficient of variation of 17%. The mean tmax was 1.22 hours. The mean terminal half-life for oral therapy was 3.32 versus 3.39 hours with intravenous therapy. Additionally the mean volume of distribution for the oral group was 19.65 and 19.71L in the intravenous group. Finally, the mean clearance for the two groups was similar with 73.16 ml/min/m2 for oral versus 72.16 ml/min/m2 for intravenous administration. The study included patients ranging up to 88 years of age and would generally apply to the older patient population.

5.8 Tamoxifen

Tamoxifen is an orally administered nonsteroidal anti-estrogen that has been available for the treatment of patients with breast cancer for many years. Absorption of tamoxifen is not well defined in humans. Tamoxifen is metabolised through CYP3A4 to a N-desmethyl metabolite. The metabolite of tamoxifen appears to contribute significantly to the pharmacological action of the agent, therefore drug interactions through the CYP system can significantly alter the formation of this active metabolite.[128] Unfortunately, relatively little information regarding the correlation of clinical response between tamoxifen and N-desmethyl-tamoxifen concentrations exists to quantify the pharmacodynamic interaction with clinical outcomes over the course of therapy. Agents known to inhibit the formation of N-desmethyl-tamoxifen include erythromycin, cyclosporin, nifedipine and diltiazem.[65] There are a number of new anti-estrogens and aromatase inhibitors entering the drug market.[128]

5.9 Alkylating Drugs

The alkylating agents are drugs that act through the covalent bonding of alkyl groups to cellular molecules. They alkylate DNA through the formation of reactive intermediates that attack nucleophilic sites. These drugs play an important role in chemotherapy in many combination regimens.[129] These drugs are particularly valuable in elderly patients because they come in oral forms (chlorambucil, melphalan, cyclophosphamide and lomustine) and have relatively little acute toxicity. Table VI shows characteristics of some of the oral alkylating drugs. There have not been specific studies of the pharmacokinetics of these drugs in elderly patients.

Table VI
figure Tab6

Oral alkylating drugs[129]

Full size image

5.10 Complementary and Alternative Medicine and Foods

Recent evidence suggests that at least one cancer patient in three uses some form of complementary and alternative medicine (CAM).[130,131] There is a growing database in the US to facilitate improved research on CAM for cancer, yet many gaps remain. Given the mounting evidence that CAM treatments are biologically active as well as widely used and are likely to affect drug metabolism, CAM research may affect cancer outcomes. There have been reports of interactions between the commonly used herbal remedy, St. John’s wort (Hypericum perforatum), and other medications. St. John’s wort lowers the AUC of digoxin by induction of the P-glycoprotein drug transporter.[132] It may also decrease the blood levels of indinavir and cyclosporine by induction of the 3A4 isoform of the CYP enzyme system.[133,134]

It is well documented that grapefruit juice is a potent inhibitor of CYP.[135] Conversely, cruciferous vegetables such as broccoli, cabbage and Brussels sprouts are CYP inducers. Similarly, charcoal-grilled beef, red wine, alcohol (ethanol) and cigarette smoke also induce the CYP system and have the potential to alter the rate at which many drugs are metabolised.[136]

6. Conclusion

There is increasing recognition of the importance of geriatric oncology. The elderly are the largest group of patients for the medical oncologist. Many of the drugs which have been recently approved have an improved therapeutic index for the elderly as well as a broad range of activity. This has particularly affected the treatment of patients with solid tumours such as lung, bladder, prostate and breast cancers. The introduction of newer agents for colorectal cancer will certainly have a significant impact, particularly the oral medications. This will allow a broader spectrum of patients to derive benefit from chemotherapy, particularly those with a poorer performance status. The elderly are still underrepresented in clinical trials.[137] More studies are needed regarding toxicity, drug metabolism and drug effect. Also, we will need improved ways to guide our decision making as to the appropriate therapy for this group, taking into account comorbidity, performance status and geriatric functional assessment.[138]