Cellular senescence is a program that many cells activate in response to damage and stress1. Senescence is characterized by a stable cell cycle arrest and a robust proinflammatory secretome2,3, which underlie the two major biological functions of senescence, namely, cancer protection and tissue repair, respectively4,5. Postmitotic, terminally differentiated and other nonproliferative cells can also undergo senescence when damaged, in which case their main distinctive feature is their inflammatory secretome6. In healthy, young individuals, senescent cells are efficiently cleared by the immune system in a process concomitant with tissue repair. However, during aging and in many chronic pathological processes, senescent cells are not efficiently cleared and sustain. The long-term persistence of senescent cells that escape immune-clearance mechanisms is pernicious. In this issue, Xu et al.7 show that in old mice, or mice ‘aged’ with senescent cells, drugs targeting senescent cells (known as senolytic drugs) increase remaining lifespan by 36%, enhance healthspan, reduce frailty and delay age-related diseases.

It has recently been shown that many human pathologies are associated with the presence of persistent senescent cells4 and that selective killing of these cells has a remarkable therapeutic impact across multiple mouse disease models2. A class of pharmacological compounds that preferentially kill senescent cells have been discovered and are known as ‘senolytic’ drugs2. This type of drug has shown therapeutic activity in a surprising variety of mouse models of human disease, including pulmonary fibrosis, atherosclerosis, osteoarthritis and chronic kidney disease2. Regarding aging, a previous study has shown that the continued elimination of senescent cells in mice, from midlife (1 year of age) until death (which occurs from 2–3 years of age), delays aging-related pathologies, such as kidney and heart deterioration, and extends the lifespan of these mice by ~25% on average8. However, despite their importance, these studies on aging were carried out using sophisticated genetic manipulations that are difficult to translate to humans.

In their study, Xu et al.7 first developed a novel experimental setup that recapitulated ‘aging-like senescence’ in young mice1. The authors then generated senescent cells damaged in vitro by either irradiation or chemotherapeutic drugs and injected these intraperitoneally into young mice. They found that this triggers loss of physical performance, which is a very common trait of human aging. Therefore, senescent cells, even in small ratios, can contribute significantly to tissue dysfunction (Fig. 1). This model is a welcome advance that circumvents the need to perform lengthy and costly experiments with naturally aged mice and will hopefully expedite research in this area.

Fig. 1: The senolytic effect.
figure 1

Aging is accompanied by a progressive increase in the number of senescent cells in aging tissue (left). Xu et al.7 show that injection of senescent cells to young animals (top) can recapitulate some aging phenotypes, particularly that of reduced physical performance. Elimination of these cells with senolytic compounds, such as the combination dasatinib–quercetin, has remarkable beneficial effects on mice (bottom). Treatment of naturally aged mice with the same senolytic compounds also improves physical performance in old mice.

Full size image

The authors then studied the effect of a prolonged senolytic treatment on their aging-like senescence model as well as in naturally aged mice7. They have previously characterized the senolytic treatment, consisting of a combination of dasatinib and quercetin9, used in this study both in vitro in senescent cells and also in several disease models in mice (Fig. 1). These are Food and Drug Administration (FDA)-approved compounds for treatment of a cancer indication with broad-spectrum kinase inhibitory activity. A single 5-day oral treatment with dasatinib–quercetin was sufficient in the aging-like senescence model for the mice to recover physical performance equivalent to that of untreated mice7. In the case of naturally old mice (20 months of age), biweekly administration of dasatinib–quercetin over 4 months was sufficient to improve their physical performance compared to untreated mice (measured at the end of the treatment) and to detectably extend their average longevity, albeit moderately (~5% on average, from a median longevity of 937 to 996 days)7. Of note, the dosing used did not produce noticeable toxic effects, such as pulmonary edema, which is a major limiting toxicity of dasatinib alone.

There are several challenges in moving these drugs into the clinic as a broad treatment for aging. First, there is evidence that different senolytic compounds target different senescent cell types, and so it is likely that in order to treat different diseases of aging, the right markers of senescence for each pathology or aged tissue will need to be identified10.

Based on the authors’ model of senescence described above, it will be possible to determine the number or ratio of senescent cells that needs to be removed in old organisms to achieve functional benefits. Notably, in the authors’ model, injection of 0.5-million senescent cells was sufficient to reduce the fitness of young healthy mice, and this represents a tiny fraction of cells within the organism. Senescent cells, however, are not evenly distributed in old organisms. Only recently are methods being developed to quantitatively determine the number of senescent cells in animal tissues11. In adipose tissue, the amount of senescent cells in old mice may reach up to 14% of the cells, whereas in intestine, spleen and lymph nodes, ~3% of the cells are senescent11. In humans, enabling the assessment of the mass of senescent cells in a tissue will help to determine the best age to start a senolytic treatment, taking into account that biological and chronological ages may not coincide.

However, it is most likely that the initial proof of concept that senolytics can be applied in the clinic will be in clinical studies in which senolytics are administered to the elderly before they develop severe disabilities. For example, it might be that days or even weeks before undergoing elective procedures (such as elective hip replacement), senolytic therapy will be administered, and clinical progress will be compared between individuals taking senolytics and those with no treatment. The data from the current paper suggest that resilience of the elderly during such an event will increase after such treatment and that their recovery will be better and disability will be prevented12. With success, indications could be broadened.

The elimination of senescent cells with pharmacological strategies and through use of FDA-approved drugs is a major advance that may be readily tested in aging humans. These are exciting times that mark the coming of age of decades of previous research on cellular senescence.