The UKTOCS ovarian cancer screening trial presented its impact of screening on cancer mortality results this morning at the Royal College of Obstetricians and Gynaecologists in London. They live streamed the event, so I was able to dip in and out as the meeting took place.
While not definitive, the results certainly suggest that screening may well reduce mortality. Another three years will be required to confirm the presence of a late effect of reduced mortality. In the words of Professor Ian Jacobs, who chaired the meeting, there is "encouraging evidence of mortality reduction".
Diagnosis at an earlier stage was shown and may be encouraging, but it is really only a breakthrough if in practice that leads to a reduction in the number of women who die from the disease. Otherwise the harms of screening (which include complications from what turns out to be unnecessary surgery) will not outweigh the benefits.
The meeting heard from, amongst others, patient Sarah Smith, Professor Usha Menon on the background, design and implementation of the trial, Professor Lesley Fallowfield on the psychological outcomes and Professor Max Parmar and Dr Steve Skates on the mortality results.
The background information to this trial includes types of screening, ultrasound only or multi-modal and the personal CA125 algorithm for each woman rather than using an all purpose cut-off point.
Professor Fallowfield presented data on anxiety, the effects on the lives of the participants (especially those recalled for additional testing) and those who left the trial. This presentation was particularly interesting for me as I could relate it to my own feelings while in the 'sister study' UKFOCSS (which has yet to report on the second phase in which I took part).
The mortality results were presented in several ways, with the multi-modal arm showing a reduction in mortality of 15% overall. There were clear differences in results for 0 to 7 years and 7 to 14 years of screening, which chimed with pervious studies showing that any benefits were seen in the long rather than short term. One presentation of results excluded prevalent cases by looking at the point at which levels of CA125 rose in order to exclude those where ovarian cancer developed prior to entry into the trial.
While there is no definitive evidence that screening saves lives, there is evidence of a late effect and this reflects what was found in the earlier PSA marker trial for prostate cancer. It is consistent with a finding in UKCTOCS that there was an 8-9 year period for ovarian cancer deaths to show up in the control arm and the finding that the curves are still diverging. We need another three years of follow up to observe this late effect and therefore we are still in a period of uncertainty.
In his summing up, Professor Jacobs pointed out that the trial started with some pieces of the jigsaw in place, while we can say that the trial has now been able to add some others such as a high level of compliance, sensitivity and a shift in diagnosed disease stage. The pieces relating to reduced mortality and cost effectiveness are still to fall into place and complete the picture.
Both Professor Menon and Professor Jacobs paid tribute to the trial participants with Professor Menon making the point that it cost the women in both time and money, as the trial was unable to pay the expenses (travel, parking costs) of attending sometimes late running appointments. Professor Jacobs closed the meeting by inviting applause for the participants.
So the conclusion is that we are not yet in a place where population screening for ovarian cancer can be recommended, but that day may come. A bit disappointing not to have a recommendation but, in view of the benefit v risk controversies that have arisen in relation to breast screening, it is better to get things on a firm footing at the outset. Given that it has taken 30 years to get to this point, it is worth waiting another 3 for a definitive answer.
CRUK's Scienceblog has an excellent graphic indicating where we are with this at the present.