èƵ

Circadian rhythms boost cancer therapies

Tumours shrink faster if given in the sweet spot in a patient's circadian cycle

YOU use a clock to decide when to get up and when to go to bed. Now it seems that timing is everything when it comes to taking cancer medication as well. The ability to pinpoint when chemo and radiotherapies will be least harmful, and most effective, should allow doctors to maximise tumour shrinkage with a minimum of side effects.

The idea of “chronotherapy” is not new – doctors have known for some time that the side effects of drugs such as oxaliplatin are less if patients are treated at night, rather than in the morning. But chronotherapy can’t go mainsteam without large-scale studies on people. “There’s enormous resistance from the medical community who say that these are only marginal effects,” says , who studies circadian rhythms at the University of Oxford. The logistics of giving patients drugs at specific times of day has also put doctors off.

Now, the development of – intravenous drips programmed to deliver drugs automatically at any time of day – is making it easier to conduct large-scale studies, and the results suggest that the effects of carefully timed drugs could be far from negligible.

Circadian rhythms affect the expression of almost every gene in our bodies (èƵ, 26 June 2007, p 8), so it is not surprising that drugs are metabolised differently according to the time of day (see diagram). “The interaction of the drug with its target will depend on the time when the cell is exposed to it,” says chronotherapist Francis L�vi of the in Villejuif.

A guide to circadian rythms

In particular, cancer cells are a prime target for chronotherapy. In healthy cells, the circadian clock acts as a pacemaker, regulating the cell’s “cycle”, which in turn governs cell division. In cancer cells, this dependence is disrupted and cell division runs rampant.

While chemotherapy often targets this rampant division, rapidly dividing healthy cells can get damaged in the process, including those lining the mouth and throat, causing painful side effects. With chronotherapy the idea is to find a kind of sweet spot, and give the drugs when the healthy cells are in the less active part of their circadian cycle, but the cancer cells are still dividing.

This is exactly what L�vi has been doing. He recently gave intravenous drugs to 114 people with metastatic colorectal cancer, and 45 with non-small-cell lung cancer over a 12-hour period. He staggered the peak delivery for different patients and found variations in the incidence of side effects. For 5-fluorouracil-leucovorin, the best times to avoid side effects were 1 am or 4 am, for oxaliplatin they were 1 pm or 4 pm and for carboplatin 4 pm was best. Side effects were up to five times as frequent during the worst times compared with the best times. The drugs were also more than twice as effective at shrinking the tumour at these times. The research was presented at an in San Francisco earlier this year.

“We really are different creatures at 4 am and 4 pm, and we need to take this into account,” says Foster, who was not involved in the work.

Patients don’t only benefit from careful timing of their drugs, however. Georg Bjarnason of the University of Toronto, Canada, has found differences in side effects depending on when radiotherapy is given. If they receive radiotherapy in the morning rather than in the late afternoon, patients treated with high-dose radiation for head and neck cancer have a lower risk of mucositis – damage to the lining of the mouth and throat that can be so painful they can’t swallow, says Bjarnason. The work will appear in a future issue of the .

And as we gain more knowledge about the cell cycle and the molecular targets that drugs hit, it is also becoming easier to predict when specific drugs will be most effective and cause the least side effects. David Orrell of the company in Oxford, UK, and his colleagues have developed a computer model of the cell cycle that allows them to simulate the effects and efficacy of drugs at different times of the day. It can even be adjusted to represent different cells in the body. “The cell cycle is an incredibly complicated thing, but our model captures the main proteins that are known to be involved in it,” says Orrell.

“It is becoming easier to predict when specific drugs will be most effective and cause the least side effects”

Using this “virtual cell”, Orrell’s team has probed the effects of , an experimental cancer drug that works by inhibiting some of the proteins involved in driving the cell cycle and cell division. The model suggests that if the drug is given three hours after daybreak, it should have no effect on normal cell division, but if given 19 hours after sunrise, it will stop normal cells dividing, causing unwanted side effects. However, because cancer cells are not under circadian control, seliciclib would affect them at all times of day.

“The ultimate aim is a computer program that would enable you to predict the best time to give a drug in humans,” says Orrell, who presented his results at the in Gothenberg, Sweden, this week.

The current model is based on mouse cells, but Physiomics is now adapting its model to reflect the human circadian system, which is around 12 hours ahead.

By tweaking the model, or combining it with a different computer model – such as one for cell death – Physiomics hopes to predict the effect of other cancer drugs, and possibly those for different diseases, such as rheumatoid arthritis, heart disease or Alzheimer’s, where circadian rhythms are likely to affect drug metabolism.

Because circadian cycles vary, tests may be needed to determine where in the circadian cycle an individual is – perhaps by taking a mouth swab or blood sample and looking for biomarkers. Drug therapy can then be tailored to the patient. “Circadian cycles do vary from person to person,” says Orrell. “The ultimate aim is to provide personalised schedules which would take such factors into account.”

Topics: Cancer