Breast Cancer - Recent Studies with Aspirin

There are more than 40,000 new cases of breast cancer each year in the UK, with 13,000 deaths. Its incidence is rising because of lifestyle factors. As with colon and cervical cancer, he said, in breast cancer COX-2 rises and increases prostaglandin activity, which then causes more COX-2 induction and synergism with growth factors in a vicious cycle leading to uncontrollable growth. In breast cells. In breast HER2 cells there is increased growth factor (GF) activity.

Angiogenesis, apoptosis, EGFR modulation and aromatase modulation. The breast is made of ducts and lobules. After pregnancy it becomes more differentiated and bigger so that pregnancy at a young age protects against breast cancer. When a breast does develop a cancer, it progresses in stages from having normal lobules to hyperplasia, atypia, carcinoma in situ, and then to invasion. COX-2 expression is very low in the first stage. Its expression is not known in the next two stages, but it is present in 60-80 per cent of in situ cancers and in 40% of invasive stages. So are COX-2 inhibitors as useful in human breast cancer as in bowel cancer?

In more than 15 animal studies the size and number of breast cancers were reduced by approximately 50% by NSAIDs and COX-2 inhibitors including aspirin, flurbiprofen, indomethacin and celecoxib. COX-2 has been shown to increase abnormal DNA adducts (the first stage of change towards cancer). High fat diets lead to over-expression of another cancer-promoting gene, HER-2, in human breast cancers: if this is over-expressed in mice COX-2 inhibitors will inhibit them.

There have been several studies on the association between NSAID use (mainly aspirin) and breast cancer, not all of which showed an effect. The Nurses Health Study conducted over many years in thousands of women showed no effects of NSAIDs in reducing breast cancer incidence, perhaps because many of those allocated to no aspirin use may have taken it against instructions. Other studies have reductions in breast cancer incidence. The results are mixed, and not nearly so clear as those for colon cancer. A meta-analysis of all the studies (6 cohort and 8 case control) found a combined relative risk of 0.82, (a reduction in risk of 18 percent) which was significant.

However, in November 2003 the Women's Health Initiative Study (WHI) was published in Cancer Research. This was a study of 80,741 women aged 50 - 79. Women who had taken two or more tablets per week for 5-9 years had a relative risk (RR) of 0.79 (CI 0.60-1.04). Regular use for 10 or more years also produced an RR of 0.79 (CI 0.56-0.91). There was a highly significant trend showing fewer cases of breast cancer with longer aspirin use (p<0.01). For ibuprofen the RR was 0.51 (CI 0.28-0.96). Neither paracetamol nor low dose aspirin (under 100mg per day) had any effect on breast cancer risk.

The dose therefore matters: women may need to take more aspirin than the single low dose daily (75mg) to prevent breast cancer. The longer they take it the lower is their relative risk of developing breast cancer.

One enzyme that may matter in breast cancer is aromatase. COX-2 induced prostaglandin production results in increased aromatase in human tumours. Aromatase converts androgens from adrenal glands into oestrogens which may themselves cause breast cancer. Blocking both COX-2 with a COX-2 inhibitor and aromatase (by using aromatase inhibitors such as exemestane or arimidase) may be helpful. A trial of this is going on in Canada.

Virchow said in the 19th century that cancer is a wound that doesn't heal. The aim is to heal the wound. Blocking COX in the breast may be a useful strategy for doing this. Epidemiological studies such as WHI are very important, indicating an effect of COX-2 inhibition. But randomised trials are still needed.

Dr Richard Sullivan, Head of Clinical Programmes, Cancer Research UK (CRUK), London, spoke on the need for research support internationally, and on progress in cancer control.