This site is intended for health professionals only!
Matti S Aapro
International Society of Geriatric Oncology (SIOG)
Multidisciplinary Oncology Institute
Clinique de Genolier
Evidence-based treatment of cancer in elderly patients is one of the key therapeutic areas that urgently needs development. Elderly people who present with cancer suffer from and die from it because of inadequate treatments.(1,2) It is paradoxical, although historically understandable, that paediatric oncology, which represents numerically so few patients, has such a pre-eminence, while more than 50% of cancer patients are in fact excluded from prospective clinical trials.(3)
Elderly patients have modified body functions related to age (physiological changes) and concomitant diseases (comorbidity).(4) Treatment decisions will also have to take into account the patient’s perceived balance between expected side-effects and benefit (in terms of quality of life and life expectancy .(5)
Physiology of ageing and specific toxicities
With advancing age, starting soon after adolescence, many body functions start to become altered. This progressive deterioration has been well documented, but its effect on a healthy person is usually irrelevant until a very advanced age, and many elderly persons have less function loss than traditionally reported.(6) However, apparently excellent general condition can mask some progressive loss of organ function. Any disease or stress can alter dramatically the delicate balance between existing functional reserves and normal physiological functioning. Of particular importance for the discussion of appropriate use of drugs in cancer therapy are changes in renal and hepatic function, as well as modifications of lean body mass and bone marrow reserves.
Glomerular filtration rate decreases with age, and several cytotoxic agents are dependent on this function. Serum creatinine values are not a good way to evaluate this loss of function, because with advancing age the majority of people have a loss of lean body mass, leading to decreased creatinine production and falsely normal creatinine serum levels. Calculating the actual creatinine clearance in a particular patient using, for example, the formula of Cockroft and Gault is more reliable in this population than an evaluation based on standard 24-hour urine collections. However, this formula is not as reliable in patients over the age of 75 and tends to underestimate renal function.(7)
Hepatic function is modified by ageing in several ways: decreased blood flow, decreased albumin production and decreased cytochrome P450 function.(8) Elderly people are often using several drugs concomitantly, and this polypharmacy may lead to possibly clinically significant changes in cytochrome P450 function. Classic examples of such changes are the induction of cytochrome P450 by phenobarbital and sex steroids and its inhibition by cimetidine. Phase I reaction modifications will have an unpredictable effect on cytotoxic agents such as the oxazaphosphorines (cyclophosphamide, ifosfamide), which are activated and also deactivated by this mechanism. Some important interactions(9,10) with cytotoxic agents are shown in Table 1.
Drug distribution and absorption
Modifications of lean body mass in elderly people are characterised by a proportional increase in fat, which increases from 15 to 30% of body weight, with a decrease in intracellular water.(11) This implies that drugs which are more hydrophilic will have a decreased volume of distribution and a higher transitory peak plasma concentration (a possible cause of anthracycline-related cardiac toxicity), and drugs which are lipophilic have an increased volume of distribution, lower peak plasma concentration and possibly increased terminal half-life. The clinical consequences of these changes are unclear, as in part they may be counterbalanced by the relative hypoalbuminaemia of elderly people, which in turn leads to increased liver extraction fractions. Some cytotoxic agents, such as etoposide and taxanes, are highly protein-bound and can potentially also be displaced, with resulting higher concentration of free drug, by concomitant use of salicylates or probenecid. It should be mentioned that there are conflicting data about changes in drug absorption after oral administration in elderly patients, and that the commonly cited decrease of gastric acid production is probably not true in patients without comorbidity.(8) However, atrophic gastritis is a common condition in elderly people above the age of 80 and may lead to decreased absorption of folates, etoposide and several other orally administered agents.(12)
Bone marrow reserves
Many cytotoxic agents are myelotoxic, and unpredictable myelotoxicity can arise in elderly people, even if doses are adjusted to account for differences in pharmacokinetics. Such observations are, however, mostly limited to frail (malnourished) patients.(13)
Finally, as many cytotoxic agents (vinca alkaloids, epipodophyllotoxins, taxanes and platinum derivatives) are neurotoxic, it is important to realise that an elderly person may be considerably handicapped by the loss of peripheral sensitivity, as well as by ototoxicity, which may lead to clinically significant hearing loss.(14) Several ways to prevent or reduce this neurotoxicity have been suggested, including the use of glutamic acid against vincristine induced toxicity.(15) This suggestion may not be ideal in view of the possible toxicity of glutamate.(16) Elderly people can also have more central nervous system toxicities, and this is of prime importance when one considers the use of fludarabine.(17)
Principles of drug dosing
Liver function abnormalities
In the presence of liver function abnormalities, which lead to an increased bilirubin level, several drugs have to have a dose adjustment. This is not different in elderly or younger patients, although in both situations there is a paucity of published data so it is necessary to rely instead on theoretical considerations.
Antimetabolites, epipodophyllotoxins (if renal function is normal) and most alkylating agents do not seem to be more toxic in these cases, but doses of anthracyclines and anthraquinones (except idarubicin) should be reduced by 50% if bilirubin is above 1.5mg/dl and by 75% if bilirubin is above 3.0mg/dl.
The reason why idarubicin does not need adaptation in case of increased bilirubin values is related to the renal excretion of the idarubicinol active metabolite.(18,19) This means that idarubicinol does, however, need careful dose adaptation in case of renal insufficiency, and these guidelines are not yet available.
The 4-deacetyl metabolite of vinblastine is cytotoxic, and in case of impaired hepatic function the same guidelines are suggested as for anthracyclines. Taxanes also undergo hepatic metabolisation and should be managed according to the same guidelines as anthracyclines. For docetaxel there is a recommendation to use it at 75% of the dose if there is a simultaneous abnormality of both alkaline phosphatase and transaminase values, even in the presence of normal bilirubin levels.
Renal function abnormalities
The Cockroft and Gault formula has been shown to be a reliable method of measuring renal function in patients up to the age of 75.(20) Above that age one may want to evaluate, in patients with obvious changes in body mass composition, the actual urinary creatinine output.
Several formulae exist that can then help the clinician to adjust the drug dose to the level of creatinine clearance, especially for carboplatin.(21) The hepatic metabolism of taxanes, doxorubicin, epirubicin and vinca alkaloids in principle permits their use at full dose, even if renal function is impaired.(22) A formula based on the fraction of active drug that is renally excreted allows one to adapt the dose of most cytotoxic agents with such characteristics.(23) Epipodophyllotoxins can probably be used at full dose until a clearance of less than 25–30ml/min, if liver function is normal. Bleomycin, carmustine, cisplatin, 2-CDA, camptothecin derivatives, cytarabine, fludarabine, ifosfamide and other agents need to be dose reduced as soon as the creatine clearance is below 60ml/min, a commonly accepted limit. There are unfortunately very few published studies of these agents in renally impaired patients, and none in patients above the age of 75 with renal impairment. It is also important to note that patients above 70 are not different from younger patients when evaluated for cisplatin-induced nephrotoxicity,(24) and this has also been shown to be true for patients above the age of 80.(25) Thus, if thrombocytopenia is of concern, one can, in patients who can tolerate a careful hyperhydration, use cisplatin instead of carboplatin in elderly patients.
Use of growth factors to prevent myelotoxicity is recommended in all elderly patients at risk of febrile neutropenia.(26)
Elderly patients do not age in a uniform way, and several parameters have to be evaluated before one decides the best therapeutic approach. Until appropriate studies are available, the clinician can use existing single drugs and drug combinations, depending on the palliative or potentially curative intent of the intended treatment, applying the existing guidelines discussed above.