CCG Carpenter Breastcancer H-Mini

CCG_Carpenter_BreastCancer_H-mini

Speaker key

RORob Carpenter

ROHi, I’m Rob Carpenter. We’re going to talk tonight about early diagnosis of breast cancer. Of course that involves breast screening, and so I want to talk about some of the more GP-oriented aspects of breast screening tonight; a little bit about genetic variance; and I have a case-study on a BRCA invitation positive patient that I’d like to work through with you. And then I thought it would be sensible to go through not who you should send but who, maybe, you could manage in general practice safely as just a last, small part of the talk.

So, breast screening, as you might now know, has been extended out to 47-73, and that should be rolled in by this year. It still involves mammography every three years, and regardless of what the continuing debate dictates, there is a survival benefit, it’s significant. And in relation to the main disadvantage that’s suggested, basically what it breaks down to, in a meta-analysis, is two lives saved for every one breast cancer over-diagnosed. And what I mean by that is cancers that otherwise would not have disclosed themselves during a woman’s lifetime.

Whatever you might think, and whatever your view about the survival benefits, this is a very high-quality service. It’s probably the most audited medical process we have, and the knock-on effect of all of that is not just the quality that we can guarantee in the service but the effect it’s had on the symptomatic service, which is still by far the largest source of breast cancer in the UK at the moment.

Well, let’s look at the disadvantages. And having said that there is a survival benefit I think that we’ve generally underplayed the disadvantages in the past, and I think more fairly now these should be discussed with our patients. There are inevitably biopsies for benign disease: to get the diagnostic accuracy right you sometimes have to over-call, particularly in the first mammographic screening round. This still represents a very small minority, though, of interventions.

For those women where the cancer is detected early but they don’t achieve a survival benefit, all you’ve done is to extend the period of time that they live with breast cancer, which is a significant disadvantage. The excess cancers represent some ductal carcinoma in-situ and some early breast cancers, the small tubular cancers, for example. These all represent disadvantages of screening.

And there’s the other large group of invitation anxiety, those women who are challenged by an invitation and might never have confronted the fact they might have breast cancer represent a source of morbidity.

Identify the beast. It’s a consultant surgeon. It’s a… it’s not Dolly, but I am Welsh, remember. It’s a sheep, okay. We all agree it’s a sheep. And this. A goat, yes, sure. And next. Okay, well, you know breast… despite all of our best attempts, this group represents a significant number of breast cancers that we diagnose through screening, okay. We can’t tell whether they’re the well-behaved sheep or the badly behaved goats. And this is the problem with the cancers that we detect is our ability to predict the ones that we can ignore, if you like: treat normal or treat minimally, and those we have to treat aggressively.

They are, in fact, goats, because their tails go up, and sheep their tails go down. Okay. So now you’ve learnt something, if you’ve learnt nothing else tonight.

So this… these are tumours less than a centimetre, and these are hazards of recurrence with time from diagnosis, and you can see there’s hardly any recurrence in tumours that are a centimetre in size.

Compare that with tumours that are three centimetres in size and you can see this recurrence go. Significantly more are going to recur in their lifetime. And that’s why breast screening works: because you pick up cancers at a much earlier stage, okay. Hopefully before they’ve metastasized, and so that’s where the survival benefit accrues. And in terms of survival, you can use disease-free survival, if you like, as a surrogate. If you wait long enough then these disease-free survival curves will translate into survival for the patients.

Okay, so this line won’t surprise you. It’s from industrialised nations in 2002 showing the incidence of breast cancer, first of all, and you can see breast cancer’s only just behind colorectal cancer. But look at the survival, five year survival for the patients within this group. I mean, breast cancer survival is phenomenally good. This message isn’t out there, that with modern treatment, women with breast cancer do extraordinarily well. This isn’t breast screening patients; these are all women with breast cancer.

And that’s a good-news story we need to get out there. And that was in 2007, the five-year follow-up was in 2007, and what we can see is these mortality curves from Australia, the US and the UK. We’ve seen a decline in mortality commenced about the late 80s. And that’s unduly due to better systemic therapies, better surgical therapies, better targeted therapies; but also screening was introduced over that time period and has contributed to some of this decline. And the projected survival curve for the UK by 2015 is there. So again, this is solid, good evidence that we’re doing well in terms of breast cancer treatment in the UK.

So I thought that we’d come closer to home, so these are figures, now, from Central and East London Breast Screening Unit; and I was a surgical leader of the unit for many, many years. Just taking figures for 2004-2005, these are the national figures for women screened; total number of cancers detected; invasive cancers and in-situ cancers. And the first thing that screams out at you is the large number of in-situ cancers that are detected; maybe 20% in a screening population; 2% or less in a symptomatic population.

So these are the cancers that are, as well as the early invasive cancers it’s the in-situ cancers that are easily detected.

The figures for Central and East London are only half of what they should be because uptake during those years was less than 50%; and that’s something where maybe we could discuss a little bit more about why that should be, and whether we should try to influence that, in fact.

And this is why in-situ cancer is picked up: because it’s associated with microcalcification so it’s easy to detect on screening mammograms, particularly in post-menopausal women when the background density in the breast is much less.

Is there a range of behaviour within in-situ cancer? Yes, there is. This is an example of low-grade ductal carcinoma in-situ, which, although it’s associated with a relative risk of 15 above general population in 15 years, if you leave a little bit behind it’s often extensive. And so surgery is generally more extensive for this relatively well-behaved in-situ cancer, in fact; as opposed to this high-grade ductal carcinoma in-situ, where the relative risk of recurrence is 20 times general population, over a five year period, not over a 15 year period; and 50% of recurrences in this condition are invasive. And of course, if you get an invasive recurrence after treatment for in-situ cancer, you’ve converted a curable condition into a potentially incurable condition.

Here’s the rub; A, we can do these broad generalisations around behaviour of these early screen-detected cancers, but unfortunately an awful lot of women are having mastectomies for ductal carcinoma in-situ, and this slide is self-explanatory: it’s by screening unit across the country, and you can see it’s not uncommon for 40-50% of women with DCIS to get mastectomy in the UK.

So, I thought we’d move on now to the high-risk families. The BRCA group form the largest group; they’re all extraordinarily rare. And the whole group only represents 3% of all breast cancer, okay? So you’re not going to diagnose many breast cancers each in your practice. You’re certainly not going to diagnose many, unless you have a family involved you’re not going to make many diagnoses in this group. And these are the actual mutations that we can test for: most commonly BRCA and P53.

I’ve put this up because I find there’s still confusion about who GPs feel they need to refer, and I think this is the simplest, it’s actually the NICE version. I’ve got a copy here that somebody can have if they want; it’s on the NICE website but it’s pretty close and it concentrates on, for instance, second-degree relatives and whether or not there’s a male with breast cancer and whether or not there’s ovarian cancer in the first degree, second degree relatives; and also whether or not it’s bilateral or unilateral. So these are the groups to refer to the high-risk clinics or the genetic clinics.

Other high-risk groups to refer that are not familial but for those women who’ve had super-diaphragmatic radiotherapy for lymphoma; and also women who’ve had a benign biopsy in the past where we’ve found atypical hyperplasia or lobular neoplasia. They also need to be screened; they’re at higher than population risk.

This introduces problems for us, because many of these familial groups are very young. Many of the women who’ve had radiotherapy for lymphoma are very young and mammography is not very good for women with dense breasts. And so, this, inevitably, means we’ve to introduce MRI scanning. But for women in high-risk groups, younger women in high-risk groups are going to get MRIs every year as part of the screening process, until they get to the post-menopausal age group.

Okay, who should be referred urgently? Everybody knows now, everybody’s seen the guidelines. A patient over 30 and these, okay; this isn’t new to you. Pain, if associated with a lump, etc; nipple discharge, all women over 50; women under 50 with these or an associated lump; nipple eczema: notice I say nipple eczema, I’m going to stress that in a minute.

Okay, look, there are a massive group of women that I see in my clinics who could be treated in general practice very safely. I can’t remember the last time I saw breast cancer in a woman complaining of cyclical breast pain, okay; or post-menopausal, non-cyclical breast pain. It tends never to happen. And in the absence of any clinical signs that might make you think they have a breast cancer I think this is ideally managed, at least initially, in general practice.

I think you can be confident that you’re not going to miss things if you do that. Younger women with tender, lumpy breasts; women under 35 with non-spontaneous discharge: what I mean by that is, they don’t need to squeeze it to see it. And please don’t ask women to show you how they squeeze it because I’ve never seen such contortions to get a drop of discharge. If it’s not spontaneous, it’s not significant.All right, I’ve got 30 years of experience of doing this but I think that these are messages that we could get out to the benefit of patients and everybody else in the system.

So, this is Paget’s disease. It’s classic Paget’s disease, it only tends to affect the nipple or at least start in the centre of the nipple and then only in very late cases does it move out involve the areola. If you have a patient with a patch of eczematous change on the areola and the nipple looks normal, it’s not Paget’s disease and you can treat it like a dermatalogical condition. But beware, because this is the earliest case of Paget’s disease I’ve seen, but it conforms to the rules, okay? It starts in the centre; the areola is spared. So this would certainly be a case; if this was a new finding in a patient that’s to be referred - even something as early as that - because that’s Paget’s disease.

Acyclical breast pain, I think, is also easy to diagnose when you know that, if you examine the breast and the breasts are normal in women after the menopause, if you can then find a tender spot on the chest wall with gentle palpation; get the end of the breast, press the costochondral joints; if that reproduces the pain, you’ve got the diagnosis. And I wouldn’t refer if a woman’s had a screening mammogram within a year, for example. If you’re worried, obviously, refer, but it can be managed, I think, safely. And it’s the same with younger women with cyclical pain; if they’re under 30 and I’m happy the breasts are normal, I don’t do any investigating.

[on screen: how do you manage young women who are anxious about their breast pain?]

Well, at that point you’re going to have to refer, aren’t you? I think that’s entirely fair. I understand that. But there’s something else going on as well. Because I can examine the same patient, maybe, and in a very authoritative way, say, look, your breasts are entirely normal. AndI’m entirely happy to see any patients, but I still think that there are a group that perhaps we could manage a little bit better and safely in general practice.

I think a great way of thinking about lumps in women’s breasts is, if you’ve got a woman under the age of 30 with a breast lump, it’s almost always benign. If you have a woman over the age of 70 with a breast lump, it’s almost always cancer, okay?

But I’ve really enjoyed that.

1