Mini Review Immunity and ageing in man Graham Pawelec * University of Tu ¨ bingen Center for Medical Research, Waldho ¨ rnlestr. 22, D-72072 Tu ¨ bingen, Germany Received 7 September 2006; accepted 12 September 2006 Available online 21 November 2006 Abstract Immunosenescence resulting in decreased ability to control infectious disease contributes to morbidity and mortality not only in the very elderly, but in all likelihood already from middle age. Studying immunity in humans is therefore essential for developing treatments to restore dysregulated immune responses and assure healthy longevity. The past year has seen many significant advances in our knowl- edge of age-associated alterations to immunity in elderly people, only some of which can be briefly reviewed here. Ó 2006 Elsevier Inc. All rights reserved. Keyword: Immunosenescence 1. New data from longitudinal studies Seeking immune alterations relevant to healthy ageing can be approached by comparing groups of young and old individuals, and yield important data on age-associated changes. However, the genetic background, lifestyle history and current circumstances of groups compared in such cross-sectional studies are always a source of confounding factors in discerning the changes which are critical for healthy ageing. Another major disadvantage of cross-sec- tional studies is that they can only provide evidence of associations between certain parameters and clinical out- come, but can never identify predictive factors. The same population of individuals followed over time offers the pos- sibility of correlating baseline parameters with subsequent events on follow-up and allows testing of factors predictive of healthy ageing. Longitudinal studies are logistically and financially difficult in long-lived species like humans, but by setting baseline at an already-advanced age, studies can be carried out over a realistic length of time. Although these starting populations are selected by virtue of longevity, immune and other factors influencing survival can be iden- tified and then also tested for their relevance in younger populations. In this context, the Swedish OCTO and NONA-immune studies have been ongoing for more than a decade now and continue to yield exciting data. In the follow-up analysis published in 2005, evidence for an allo- static load contributing to mortality was presented for the first time. Thus, immunological parameters such as excess CD8 + CD27 À CD28 À T cells, low T cell proliferative responses in vitro and low IL 2 secretion predicted mortal- ity, as did high plasma IL 6 and cognitive impairment inde- pendently. However, both sets of parameters together had a supra-additive effect; hence the concept of allostatic load (Wikby et al., 2005). Up until this year, individuals moving into this previously defined ‘‘Immune Risk Profile (IRP) group had been found to remain there until they died. One major hallmark of IRP individuals is the swollen num- bers of CD8 cells in the periphery, coupled with a slight decrease in the number of CD4 cells, so that a defining fac- tor for the IRP is a CD4:8 ratio <1. However, Wikby et al. (2006) have now reported two IRP individuals who had moved out of the IRP category over the two year period between sampling. These subjects had achieved this feat by reducing the numbers of CD8 cells in the blood, possi- bly associated with high plasma levels of both IL 6 and IL 10 (usually mutually exclusive). These unique individuals did indeed experience a survival advantage over that majority of their contemporaries who had stayed in the 0531-5565/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.exger.2006.09.005 * Tel.: +49 7071 298 28; fax: +49 7071 29 55 67. E-mail address: Graham.pawelec@uni-tuebingen.de www.elsevier.com/locate/expgero Experimental Gerontology 41 (2006) 1239–1242