Thank you very much. Good morning everybody and thank you for the opportunity to present our results at this press conference. I have no direct financial relationships to disclose; my hospital receives institutional funding both for non-profit and profit clinical trials.

With the introduction of mammographic screening interepithelial neoplasia represents at least 20% of all breast neoplasms. This entity includes atypical ductal hyperplasia, ductal and lobular carcinoma in situ. These disorders have a slightly different heterogeneous behaviour. While ADH and LCIS may be offered tamoxifen based on the results of the NSABP prevention trial, with a low uptake however, there is a tendency in the last few years to de-escalate the treatment of DCIS, which includes radiotherapy plus five years of tamoxifen, because of the lack of evidence on overall mortality. We know very well that tamoxifen is very effective in prevention but its toxicity, namely rare cases of endometrial cancer, deep vein thrombosis and menopausal symptoms represent an important barrier for its broad use in that population at increased risk from breast cancer.

This trial is quite old and was developed in the ‘60s. At that time the minimal effective dose was not assessed. This drug binds to the oestrogen receptor with a saturation kinetics but it’s not clear what is the minimal active dose to elicit its biological and clinical effect. So our hypothesis was that a lower dose, namely 5mg a day, and a shorter duration of treatment, namely three years, was as effective and less toxic than 20mg a day. So some fifteen years ago we conducted a randomised window of opportunity pre-surgical trial in women waiting for surgery where we randomised women to 20mg or 5mg or 1mg per day. You can see here very clearly that the lower doses, even 1mg per day, was not inferior to 20mg in decreasing breast cancer proliferation as measured by Ki-67 whereas in the control arm there was an increase of four weeks impact.

So the study design of this trial is the following. Women aged 75 years or younger with either ADH or LCIS or ER positive or unknown DCIS were randomised to 5mg of tamoxifen per day or placebo for three years plus at least two years of follow up. The primary endpoint of the trial is the incidence of invasive breast cancer or DCIS. 500 participants were enrolled from 14 centres in Italy. The women were assessed every six months with clinical visit and quality of life questionnaires and mammography was repeated every 12 months.

The median follow up of the current analysis is 5.1 years and is based on a total of 42 primary breast cancer events. These are the main subject and tumour characteristics at baseline. You see there is no imbalance between the two arms, mean age was 54, 45% were premenopausal. As you see, the Mediterranean diet works because the mean BMI was 25, 20% had ADH, 10% had LCIS and 70% had DCIS. Two-thirds were ER positive and one-third was unknown. There was less than 10% of HER2 over-expressers. Quadrantectomy was performed in more than 80% of the women and radiotherapy was delivered to the women with high risk DCIS, namely G3, comedo, positive margins or younger age.

These are the main findings of the trial. There was a 52% reduction in the cumulative risk of developing a recurrence in the low dose tamoxifen arms – 28 versus 14 events compared with placebo. This difference is statistically significant and the annual risk of events declines from 24 per thousand per year to 12 per thousand per year. Likewise, and perhaps more impressively, there was a 75% reduction in the risk of contralateral breast cancer with low dose tamoxifen. Of course, this analysis is based only on 15 events so we have to be cautious.

Serious adverse events, of course, in this de-escalation trail toxicity is as important as efficacy. We saw one case of stage one endometrial cancer with tamoxifen, one case each of deep vein thrombosis and one case of pulmonary emboli in the placebo arm; no difference on either neoplasia or coronary disease; one death on tamoxifen, two deaths on placebo. If we compare these findings with the NSABP-P1 prevention trial with a dose of 20mg per day we would expect 2.7 endometrial cancers on tamoxifen and 2.4 DVT or pulmonary emboli.

So patient reported outcomes are as important as efficacy here. We carefully assessed the frequency and intensity of hot flushes which is the most frequent side effect of tamoxifen using a semi-annual questionnaire self-reported by the women using the method developed by Loprinzi and co-authors. There is a slight, borderline significant, increase in the number of hot flushes per day, less than one hot flush extra on tamoxifen. Whereas if we multiply the frequency by the intensity there was no significant difference between the two arms. Perhaps more importantly, vaginal dryness or pain during intercourse was not affected by the use of tamoxifen and likewise musculoskeletal pain or arthralgia, which are the most important side effects of aromatase inhibitors, were not increased by the tamoxifen.

So, I want to conclude by saying that we have shown that 5mg/day for three years can halve the recurrence of breast interepithelial neoplasia in line with the effect noted with 20mg per day or even slightly better in the NSABP B-24 trial. Low dose tamoxifen decreased contralateral breast cancer by 75%, suggesting a strong preventive potential. The rate of endometrial cancer and DVT on 5mg was no different from placebo and two times lower than 20mg. The menopausal symptoms were not worsened except for a borderline effect on hot flushes.

We think our results have external validity, given the pragmatic nature of the trial and the easy accessibility of the drug and are therefore generalizable. Because the 5mg tablets are not available in the market you can either cut the 10mg tablet or use a 10mg every other day which is applicable in clinical practice from tomorrow. The study was supported by the government, the Italian Ministry of Health and two charities – the Italian Association for Cancer Research and the Italian League Against Cancer. We are indebted to these supporters for their continuous support through non-profit trials for fair, equitable and affordable medicine. Thank you for your attention.