Mondays at 8.30am, repeated at 8.00pm
with Norman Swan
18 April 2005
An Australian study suggests that the benefits of chemotherapy have
been over-sold. Norman Swan talks to Associate Professor Graeme Morgan
who's a radiotherapist at Royal North Shore Hospital in Sydney and to
Professor Michael Boyer who's Head of Medical Oncology at the Sydney
Cancer Centre, Royal Prince Alfred Hospital.
Good morning Fran and welcome to the program. This morning on the
Health Report globalisation writ small. Why go down the road for your
surgery when you can avoid the waiting lists, get it cut price in
another country and have a holiday with the money you’ve saved? A
personal story involving new knees coming up.
And, has cancer
chemotherapy, the use of drugs to treat malignancies been oversold?
That’s the clear implication of a paper published by some Australian
cancer specialists, two of whom, perhaps non-coincidentally are
radiation oncologists – radiotherapists.
Anyway in this summary
of evidence, the assertion is that chemo has only added about 2% to
cancer survival. The lead author is Association Professor Graeme Morgan
who’s at Royal North Shore Hospital in Sydney. Is this, I wondered, an
in house battle, the revenge of the radiotherapists?
Well one could cynically say that but the reason I did it was that we
were sick and tired of hearing about these new drugs and it wasn’t
really cementing into anything. And the reason for my doing the paper
was to really show that there hasn’t been any improvement in survival,
or the improvement has been very, very modest despite all these new
drugs and new combinations and bone marrow transplants.
Norman Swan: So what did you do in this study?
Well what I did was that I took the major cancers and got their
incidence from the Australian data sets and also got the breakdown of
those cancers into their stages and also then compared that with the
data from America.
Norman Swan: So you knew how many people were coming down with cancer per 100,000 of the population or whatever?
Yes, we knew the exact number who were diagnosed in I think the years
were 1998 and we then broke it down into the various stages and we
looked at those stages what impact chemotherapy would have on a
particular stage and a particular cancer.
Norman Swan: And where did you get that information from?
Graeme Morgan: We searched the literature – what we looked at was meta-analyses on a particular cancer.
These were reviews of studies bringing together the available evidence
on a certain form of chemotherapy for a certain cancer. So it wasn’t
just one study, they’d amalgamated several studies. And did you do it
for all cancers?
Graeme Morgan: We did for 85% of the
cancers, we didn’t do it for leukaemias and the reason we didn’t do it
for leukaemia is that in acute leukaemia it’s very difficult to
differentiate between adults and children and the numbers are fairly
small. And in chronic leukaemias, particularly in chronic lymphatic
leukaemia which is basically a disease of the elderly, it’s an indolent
disease and the median survival is way over ten years. And we also
excluded smaller cancers.
Norman Swan: So there might be
a slight bias against chemotherapy because you’ve eliminated the
leukaemias which can be quite chemo sensitive.
Well yes, but the other reason that I eliminated those because they are
mostly looked after by haematologists rather than medical oncologists.
So this was really looking at the impact of medical oncology.
Norman Swan: And medical oncologists are basically physicians who specialise in cancer and use drugs.
Graeme Morgan: Yes, correct.
Norman Swan: Whereas you as a radiation oncologist do it through radiotherapy?
Graeme Morgan: Correct, yes.
Norman Swan: What were your findings?
Graeme Morgan: Well the findings were that in Australia that the five year survival due to chemotherapy was 2.1% of the total cancers.
Norman Swan: You mean the additional survival benefit?
Yeah, from the chemotherapy. So in other words if there was no
chemotherapy in Australia, the survival of all patients with cancer
would drop from 62% to 60%.
Norman Swan: You say you
allowed for a stage of diagnosis because the trend in Australia has
been to earlier and earlier diagnosis which makes 5 year survival
somewhat of an iffy figure because when you diagnose it earlier people
may survive longer.
Graeme Morgan: No not necessarily.
This is the way patients present. The other thing was that I have been
a member of a number of committees looking into radiotherapy services
in Australia and as you may not be aware, only 35% of newly diagnosed
cancer patients in NSW receive radio therapy. And that’s been flat for
the last 10 years.
Norman Swan: And what should be the figure?
Graeme Morgan: The figure has always been around 50%.
Norman Swan: That’s the percentage of people according to the best evidence who should be having it but aren’t.
Graeme Morgan: Yeah.
Norman Swan: What would the figure be if you did a meta-analysis for radiotherapy? Is that that much better than chemo?
There happened to be two papers that I’ve got and they are both fairly
recent. One is here from Prince Alfred here in Sydney and the other’s
from Westmead also in Sydney. The one from Westmead shows that there is
a survival gain with radiotherapy of 16.1%. But getting back to these
reports is that one of the things that people always mention to us is
well, people can have chemotherapy and the inference was that
chemotherapy was an equal treatment. It’s obviously not.
One of the criticisms of this study is that you pooled all cancers and
not all cancers are the same. Some have much better response rates to
chemotherapy than others and if you divided it up you would actually
get a very different picture. You’ve mixed it up, averaged, when you
can’t really average when for example if you’ve got Hodgkin’s Lymphoma
the cure rate can be up to 90% and chemo contributes significantly to
that. Whereas solid tumours are pretty low, the tumours of the lining –
why didn’t you segregate by individual tumours?
I did include cancers where chemotherapy doesn’t have any impact
because this was looking at the overall contribution of chemotherapy.
If someone is listening to this, they’ve just been diagnosed with
cancer and they’ve been told they need chemotherapy and radiotherapy.
Do they walk away and say to the doctor well look I’ll just have
radiotherapy now, don’t bother with chemo?
No, it’s not saying that radiotherapy is more appropriate. What it’s
saying is that chemotherapy is oversold. I’ve asked people what they
thought the percentage was and most of them have said
Norman Swan: You’ve done a straw poll?
Yeah, a straw poll and most of them have said oh 5%, some have said 10%
and 15% but then when you ask them about the individual cancer that’s
when the real differences emerge because most people when you ask them
about breast cancer they say oh, 10% and 15% and of course it’s 1.5%.
And it’s 1.5% because most women don’t benefit from chemotherapy in
Norman Swan: In other words you’ve got to treat quite a lot of woman for one person to survive?
Well that’s right but you see there’s no data in women over 70 for the
effect of chemotherapy. Now that’s about one third of the total
population of breast cancer women.
Norman Swan: Why has it been oversold? Are you suggesting that medical oncologists in Australia are just sort of marketing shysters or what?
Graeme Morgan: Well, that’s what happened when chemotherapy first came to this country. In the 1970s that I think the first chemotherapy
Norman Swan: But Australia’s not alone in this, I mean chemotherapy’s if you look at the journals, chemotherapy is a prominent treatment.
Graeme Morgan: Yes, that’s right.
Norman Swan: So why has it come to that?
Well I’m not really sure but it has been oversold. Well the benefit has
been sold in the incorrect way because it’s reduction of risk of
relapse, it’s not absolute survival benefit.
What you’re saying is that it isn’t a proven survival for some women
but a fair number of women have got to be treated to get that
improvement in survival.
Graeme Morgan: Yes, that’s right
yeah. And the other thing as a radiation oncologist what I’ve seen is
that there are no real guide lines for palliative chemotherapy and you
often see patients referred to you who have had a number of drugs given
when in fact none of them have done any good - in the pious hope that
something’s going to happen, that the survival is going to be increased
by two months and I think that’s part of the over-selling.
Norman Swan: How is the consumer to respond to this message?
Graeme Morgan: Well I think the consumer has to really evaluate the information they’re given.
Well let’s just test you on this. I know you didn’t do the paper on
radiotherapy, but let’s just see whether you’re that much better. I
mean what do you normally say to a woman who’s got breast cancer and
it’s reasonably early, there’s maybe only one lymph node involved,
you’ve got a reasonable expectation of long term survival and she’s
been offered radiotherapy and chemo. OK, let’s leave chemo to one side,
what’s the informed consent for radio therapy here?
The informed consent with someone who’s had the lumpectomy is that the
radiotherapy will reduce the likelihood of recurrence from an absolute
risk of 25% at ten years to less than 5% at ten years.
Norman Swan: You’re saying that one out of four women will have a recurrence without radiotherapy and that will go down to one in twenty.
Graeme Morgan: Yes, that’s right.
Norman Swan: And how many women have to have radiotherapy to get that benefit?
Graeme Morgan: Well 80% of them don’t need it do they? 75% of them don’t need it.
Norman Swan: But you don’t know which 75%.
Graeme Morgan: That’s right and that’s the problem.
Norman Swan: Where do we go from here?
The interest I’ve had in this paper from overseas has been in service
provision and so that this calls for a re-evaluation of the amount of
money that spend on chemotherapy and whether we’re getting a bang for
Norman Swan: Somebody I showed this paper to
who’s reasonably expert in meta-analysis reckons you’ve done this
rather crudely and that if you were to have done this in a journal
which is more used to meta-analysis they might have actually criticised
your technique. And when she did a back of the envelope calculation
just looking at it in a slightly different way she got 6% rather than
2%. How confident are you in those results?
Graeme Morgan: We think that this is a maximum because it includes all patients who were eligible to have the treatment.
Dr Graeme Morgan who is in charge of Radiotherapy at Royal North Shore
Hospital in Sydney. And you’re listening to the Health Report here on
ABC Radio National.
Norman Swan: Needless to say medical
oncologists, the specialists who dole out chemo aren’t taking this
criticism lying down. Here’s Associate Professor Michael Boyer, Head of
Medical Oncology at the Sydney Cancer Centre at Royal Prince Alfred
Michael Boyer: Well I’m a little puzzled by
this paper. I mean on one level it purports to show that chemotherapy
adds almost nothing to the cure of patients, on another level it’s
clear that chemotherapy results in if you just accept this all at face
value, at least 1700 being alive at five years who wouldn’t otherwise
Norman Swan: Based on the Australian figures?
Based on the Australian figures, and as a person that treats cancer
patients and looks them in the eye every day I mean what we are trying
to do as doctors is cure people and so to sort of come up with the idea
that this is in some way not worthwhile is a very bizarre approach. But
I think actually that you need to look just a little bit deeper than
those superficial figures because within this paper the argument that
is made that in some way because a modality only adds a little bit to
Norman Swan: Modality being a form of treatment.
A form of treatment adds only a little to cure, and it really ignores
the way in which the modern treatment is evolving. If you go back two
or three decades we used to treat cancer with a single type of
treatment, typically surgery. Then the concept that you could do better
in terms of survival and control of symptoms by adding additional
treatments was evolved and nowadays I think -
Norman Swan: - adjuvant treatment.
So called adjuvant treatment, and nowadays I think that most people
would agree that the best outcomes for cancer is when you have a team
of people each using their own particular treatments in the right
patient and the right time to and up with the best results. So I don’t
see this as an argument of whether chemotherapy adds a little bit, or
radiation adds a little bit, or surgery adds a lot. What this really
should be about is what is the best treatment, what is the best
sequence of treatments, what’s the right time to use those treatments.
Norman Swan: So what you just see as a bit of old fashioned turf war between the radiotherapists and the medical oncologists?
I’m a little reluctant to actually say that but to be honest I think
that this reads as though a pre-conceived conclusion was arrived at and
then the data gathered to sort of support that conclusion rather than
saying OK what’s the best way to treat these diseases, let’s look
critically at what each bit adds.
Norman Swan: But I mean a 2% additional survival does not sound impressive.
Well it doesn’t sound impressive and it’s also not correct. It’s not
correct for a number of reasons. That 2% figure is achieved by
including a whole series of diseases in which chemotherapy would never
be used. The paper itself actually states that yet they are included as
part of the denominator if you like. So if you start taking those
things out and saying well OK, how much does chemotherapy add in the
people that you might actually use it, the numbers start creeping up.
If you pull it altogether that number probably comes up to 5% or 6%, I
guess what’s important is that it doesn’t go up to 50% or 60% but we
know that and we know that these treatments are at the margin. I mean
we are adding a little bit to survival and that has been the nature of
all advances in cancer treatment that you actually add to marginal
survival rather than these huge leaps with a couple of exceptions.
other point about this is that some of the figures that they use I
believe don’t represent the most accurate and the most up to date
figures that we have available.
Norman Swan: Such as?
Well such as in head and neck cancer the contribution to chemotherapy
as it is currently used, the contribution of chemotherapy is bigger
than the 4% that is claimed in the paper. With myeloma the reference
that they use and the data they are basing this on in fact is not a
paper that compares chemotherapy and no chemotherapy. It’s a paper that
compares two different sorts of chemotherapy and finds there’s no
difference between those two sorts of chemotherapy. That’s a far cry
from saying that chemotherapy versus nothing is ineffective.
One of Graeme Morgan’s points is that there’s great hype about new
chemotherapy treatments and therefore demand that they go on the market
immediately, obviously from desperate families who think that this is
going to be answer for their loved one who’s dying of cancer. Therefore
there’s a waste of resources.
Michael Boyer: Well in this
country there is a process that drugs go through firstly to be able to
be marketed and secondly to get onto the PBS.
Norman Swan: The Pharmaceutical Benefits Scheme.
The Pharmaceutical Benefits Scheme, in other words to be subsidised by
the public purse. Now this process applies equally to anti-cancer drugs
as it does to blood pressure drugs as it does to cholesterol lowering
agents. Included in that process to my knowledge is both an evaluation
of the effectiveness of the drug and also the cost effectiveness
compared to either other similar drugs or compared to nothing if there
is no accepted drug. Now if what Dr. Morgan is saying is that that
process is wrong or in some way flawed that might be the case and maybe
it should be opened up to public debate. But that’s not really what
this paper is about and so I think the two things are slightly
Norman Swan: What about the issue of informed
consent? I mean for example if you’re a woman with early breast cancer
you know you’ve had a lumpectomy, you’ve had merely one node, I mean my
understanding is give or take is that you do a bit of radiotherapy and
the chemotherapy for most women in that situation.
Michael Boyer: Or hormonal therapy depending on the exact nature of the tumour.
Norman Swan: The survival benefit is of the order of what 10% or something like that?
It varies and without going into the details one of the other problems
of this paper is it uses absolute benefits rather than relative
benefits. So the relative benefit is about a one third reduction in
your risk of death. The absolute benefit of that
Norman Swan: But for you as an individual
Depends on how big your risk of death was to start with. In other
words, if all the things you had before you get to the chemotherapy
stage have basically cured you can’t cure more than 100%. If your risk
of dying is very high, in other words your chance of having been cured
is only 20% then clearly the absolute benefit to you is larger. I
should add that that varies for almost every permutation of tumour
size, number of lymph glands and a whole lot of other features of the
Norman Swan: But on average how many women with
early breast cancer are you having to treat with chemo for one life to
be saved, or one person to survive five or ten years? And is that a
kind of routine thing you would say to people?
We probably wouldn’t put it quite in those terms but the answer to the
question is probably in the order or 20 or 25. This is not a situation
where every person you treat will clearly be cured. Equally it’s not
true to say that because you have to 20 or 25 people to benefit one,
it’s not true to say that nobody benefits. So it’s somewhere in
between. The terms in which you put it to a patient has been actually
the subject of a good deal of research and research actually carried
out in this department looking at different ways of expressing it. And
you can express this as if we treated a 100 people like you, ten of
them would benefit from treatment or five of them or whatever the
number is. You can express it as at 5 years the number that would be
alive with treatment is x percent, and with treatment it is x plus y
percent. You can express it as a reduction in your risk of dying of the
disease. Typically I think patients prefer the if we treated 100 people
just like you type scenario but having said that there is a good deal
of variability and for different people they like a different example
Norman Swan: What about the poor second cousin
kind of thing that seems to pervade radiotherapy, they feel that
they’ve been boxed into a corner with the public thinking they are just
for palliation, an end of the road treatment?
Certainly in our cancer centre here we, the medical oncologists don’t
regard radiation oncology in that way. I mean it’s an absolutely
integral part and one of the major parts in fact of treating patients
whether that be with the intention of curing a patient, or with the
intention of palliating or improving symptoms. As I said at the outset
you can’t really separate all this out, it’s not a situation where you
either have chemotherapy or you have radiation therapy. It’s a
situation where there are good ways of treating these diseases that
often encompass several different types of treatment.
And a corollary of that is do you believe that chemotherapy has been
over-hyped which is really probably one of the drivers of this paper?
Michael Boyer: Again hype is a difficult word.
Norman Swan: You may have a better public relations company.
Well I may not but certainly the pharmaceutical industry has no
shortage of public relations people and I think one has to be
realistic. The pharmaceutical industry has a vested interested
obviously in selling drugs whether they be anti cancer drugs or any
other kind of drugs and one way is to make the target audience aware of
their existence. The other issue with this paper is that by lumping all
the diseases together I think it obscures some of the detail. The fact
is that from a patient’s perspective they are not really interested in
how much chemotherapy contributes to the cure of all patients, what
they are interested in is how much it will contribute to their
particular disease and their stage of their disease. And that number
ranges from zero in some cases up to almost 100% in other cases. So I
don’t think this paper helps from a patient’s perspective. Similarly
from a public funding, or public policy point of view, lumping
everything together is not a terribly helpful way, at least in
Australia where most drugs that are now approved and reimbursed are
really approved and reimbursed for very specific indications. So there
are lung cancer drugs that can only be prescribed to patients that have
a particular gene mutation. There are anti-cancer drugs for breast
cancer that can only be prescribed where other drugs have failed. And
that reflects the way those drugs were tested in clinical trials and it
really narrows the drugs down to be used in situations in which they
are most likely to be effective. And this approach in this paper of
lumping everything together really masks that fact.
And just finally, moving away from this debate altogether, critics in
general of the cancer treatment community say look, overall it’s been a
disappointment, that things are a bit better than they used to be but
there are isolated islands and for all the money we’re spending on
cancer research, we haven’t really cracked it yet?
I mean there is some truth in that criticism although I think sort of
behind that truth is the stark fact that if you develop cancer, any
kind of cancer in Australia today your chances of being cured are in
excess of 60% and that certainly was not the case two decades ago. Now
why is that? Well some of it is because we’re better at picking disease
up earlier. Some of it is because we have better ways of ensuring that
people get operations that they need and don’t have bad problems during
or after an operation. Some of it is because we use chemotherapy, some
of it is because we use radiation therapy. On the other hand I guess
the expectation of the community is not that 60% of people should be
cured but that 100% of people should be cured and so I sense why
there’s a feeling there’s a disappointment.
However, as each
year goes the by the number creeps up and it is the nature of most
modern medicine that we don’t have the sort of breakthrough that
journalists and the media like to talk about. What we have are
incremental gains and when you add up 10 years of incremental gain
suddenly you find that your survival has gone from 50% to 60%.
Norman Swan: Michael Boyer who’s head of Medical Oncology and the Sydney Cancer Centre.
Morgan G et al. The Contribution of Cytotoxic Chemotherapy to 5-year Survival in Adult Malignancies. Clinical Oncology (2004);16:549-560
Barton MB et al. Radiation therapy: are we getting value for money? Clin Oncol (R Coll Radiol) 1996;8(3):206
Guests on this program:
Associate Professor Graeme Morgan
Royal North Shore Hospital
Professor Michael Boyer
Head of Medical Oncology
Sydney Cancer Centre
Royal Prince Alfred Hospital
Cancer - ABC Health Library A-Z
Presenter: Norman Swan
Producer: Brigitte Seega
back to The Health Report homepage