Continuing disparities in survival rates between younger and older cancer patients in Europe. Might oral chemotherapy provide one solution to the problem?
Data released by the United Nations show that Europe is currently the world region with the highest number of older persons relative to its population and is likely to remain so until 2050 at the earliest (United Nations Population Division, 2007). Most of us are familiar with this fact and may already be contemplating the economic and healthcare costs associated with the demographic shift indicated by these figures, particularly as the majority of cancers are positively correlated with increasing physiological senescence in old age (Rubin et al., 2010). It is likely however, that fewer of us are familiar with other data in relation to cancer and the elderly which suggest that in spite of much publicised improvements in cancer prevention, screening and treatment; there is evidence of a growing survival gap between older and younger cancer patients in Europe (Quaglia et al., 2009, 2007; Janssen-Heijnen et al., 2005). This scenario is markedly different from the United States where no such age gap in cancer survival rates appears to exist (Gatta et al., 2000), and we should therefore resist the temptation to believe that such inequalities are the natural sequelae of what has sometimes erroneously been referred to as Europe's ‘demographic time bomb’ (Mullan, 2002).
A series of analyses drawn from the database of the EUROCARE Working Group published by Quaglia et al. (2009; 2007) show that elderly patients, and especially women, experience higher relative excess risks of dying from their cancer than younger patients, particularly in the first year following their diagnosis (Quaglia et al., 2007). The 2009 study compared 5 year relative survival rates and prognostic changes in older (70–84 years) and younger (55–69 years) cancer patients affected by stomach, colon, breast, cervical/uterine, ovary and prostate cancers; and all cancers combined excluding prostate and non-melanoma skin cancers in 51 European populations covered by cancer registries between 1988–1999. Relative survival rates for both sexes were again generally higher in younger or ‘middle-aged’ cohorts than in the elderly for every cancer site, and the difference was again more pronounced in women than men (Quaglia et al., 2009).
The studies demonstrate that in spite of significant survival improvements for all cancers except cervical cancer between 1988 and 1999, relative survival rates (particularly one year after diagnosis) increased more slowly in the elderly so that existing survival gaps between younger and older cancer patients prior to this period have widened at the same time that many cancer therapies have improved. Importantly however, conditional survival analysis indicated that older cancer patients who managed to survive one year post-diagnosis subsequently experienced similar prognoses to younger patients, with most of the survival gap being accounting for by differential mortality within the crucial first year.
The authors suggest in relation to breast cancer, that older women are likely to present themselves belatedly with larger tumours and disease positive lymph nodes, but are less likely to be offered adjuvant therapy than younger women, even after adjusting for comorbidities or other age-related treatment considerations so that they subsequently receive ‘non-standard management’ (Lavelle et al., 2007). It would seem therefore, that advances in the management of breast cancer amongst younger women are not being replicated in older cohorts (Porock et al., 2009), a small gap in survival rates between the two age groups observed prior to the development of many treatment advances in the 1980s and 1990s becoming more pronounced in the later period covered by Quaglia et al.'s analysis (1997 to 1999). In parallel with this, it would appear that elderly colorectal cancer patients over the age of 70 continue to be under-treated, and are also under-represented in clinical drug trials in spite of the fact that their exclusion from such research is unwarranted (Rougier et al., 2004). Similar attitudes have been demonstrated by medical oncologists and radiotherapists in respect of elderly breast cancer patients (Protière et al., 2009; Hershman et al., 2008) so that their management is sometimes sub-optimal and predicated on clinical trials carried out on younger patients whose responses to specific treatment dosages may be very different (Ershler, 2006; Porock et al., 2009).
These findings should concern all of us since most cancers, including breast and colorectal malignancies are more prevalent in older people, and their incidence is likely to increase as the population ages. Quaglia et al. point out that the impact of new health policies, treatments and initiatives introduced during the last decade will not become apparent until future survival analyses are undertaken. It is for the reader then, to ponder whether the disparities in cancer survival between younger and older cancer patients identified in these studies will correct themselves over time, and consider the factors likely to influence the closure of this survival gap. In terms of therapy, perhaps the most significant change occurring in the last ten years has been the advent of oral chemotherapy drugs, most notably capecitabine as a monotherapy or in combination with other agents for the treatment of these and other cancers, although these too have predominantly been tested in younger cohorts and their results then extrapolated to older patients.
Oral chemotherapy drugs do have proven efficacy in the elderly however, and are undoubtedly beneficial since they spare them the inconvenience, discomfort and iatrogenic risks associated with hospitalisation or repeated visits to the outpatient clinic for parenteral therapy (Rousseau et al., 2010; Kotsori et al., 2010; Yap et al., 2007; Ershler, 2006). However, as Brearley et al. (2010) point out in this issue of the journal, the administration of oral chemotherapy in the patient's own home is not without its challenges. Treatment related toxicities and occasional treatment related deaths do occur (National Patient Safety Agency, 2008; O'Shaughnessy et al., 2001), although older patients usually commence treatment on a decreased dose and/or subsequently have their dosage reduced due to treatment related toxicities. The need for appropriate prescribing, dose modification guidelines, careful monitoring, assessment and reporting of adverse events is crucial therefore, and I am delighted to see these and other issues related to the management of oral chemotherapy addressed in the supplement accompanying this issue of the Journal (Irshad and Maisey, 2010; Harrold, 2010).
One of the most interesting things apparent in the oral chemotherapy literature is the finding that oral chemotherapy drugs, including capecitabine, continue to demonstrate survival benefits even when dose reductions are necessary as they frequently are in the elderly. Ershler (2006) reports on the development of one Swiss trial (SWS-SAKK-25/99) which includes a dose-finding study as well as the elicitation of efficacy and tolerability data in older women treated for metastatic breast cancer although its results appear not to have been published as yet. Yap et al. (2007) have also raised the question of ‘how low can you go?’ in relation to the use of capecitabine as a first-line treatment for metastatic breast cancer in their retrospective audit of patients receiving twice daily capecitabine at 1000 mg/m2, and there is clearly a suggestion that we might be able to treat older patients on even lower, longer regimens than we do at present (Kotsori et al., 2010), thus mitigating many of the side effects reported in this issue of the Journal and elsewhere (Brearley et al., 2010; Harrold, 2010).
The results of one French multicentre pilot study of oral capecitabine and vinorelbine in older patients with advanced breast, prostate and lung cancer (Rousseau et al., 2010) do raise one caveat however, in that their data showed a trend towards improved complete or partial response and disease stabilisation rates in patients receiving escalated doses from the fourth cycle onwards, although these patients also experienced a greater number of adverse events such as palmar-plantar erythrodysesthesia which may have additional consequences for the elderly. Further research into the dose-response and toxicity curves for oral chemotherapy agents in older adults is clearly needed therefore, and it is hoped that the results of larger clinical trials specifically focused upon their use in the elderly will soon be forthcoming. Efficacy and safety should not be the only endpoints of such studies however, since there is clearly a need for greater consideration of quality of life as a secondary outcome in such research, and the specific needs of older patients receiving oral therapy need greater elaboration, as do those of informal carers supporting them through such treatment.
In spite of these caveats, preliminary evidence suggesting that dose reductions do not significantly affect survival outcome in older patients should be investigated further in a series of well powered studies looking at different cohorts of patients since the possibility that moderate doses of oral chemotherapy agents have the capacity to improve survival rates in older patients affected by a range of common cancers whilst limiting known toxicities is too good an opportunity to miss. It may also go some way towards improving current disparities in survival rates between younger and older cancer patients in Europe and elsewhere. Meanwhile, strategies aimed at improving clinical presentation and referral times for older cancer patients should be developed, and attempts made to eradicate the perception that late presentation or the suboptimal treatment of older cancer patients are acceptable or inevitable consequences of advanced chronological age. ‘Optimal treatment’ may indeed mean ‘less’ in this instance, but the benefits of developing a much better evidence base for the prescribing of these drugs in the elderly is clearly apparent.
