AstraZeneca PLC (AZN) 2003 Q1 法說會逐字稿

Good day ladies and gentlemen and welcome to the AstraZeneca Q1 Results Conference Call.

This call is being recorded.

Today's Question and Answer Session will be conducted electronically.

If you would like to ask a question, please press the star or asterix key, followed by the digit '1' on your touchtone phone.

At this time I would like to turn the call over Mr. Ed Seage (ph) with AstraZeneca.

Please go ahead sir.

Thank you operator.

Welcome ladies and gentlemen.

Joining me on the call today from London is AstraZeneca's Chief Financial Officer, Jon Symonds(ph), Mike Anse(ph) from Corporate Affairs and of course the rest of the Investor Relations Team.

Before I hand over to Jon, I would like to read the following statement.

The company intends to utilize the safe harbor provisions of the United States Private Securities Litigation Reform Act of 1995.

Participants on this call may make forward-looking statements with respect to the operations and financial performance of AstraZeneca.

By their nature forward-looking statements about risk and uncertainty, results may differ materially from that expressed or implied by these forward-looking statements.

The company undertakes no obligation to update forward-looking statements.

Now over to Jon Symonds.

Jon.

Thanks Ed.

Good afternoon everyone.

You have had the press release for a few hours now and I know today is a busy day.

I will keep my remarks focused on the key highlights to allow plenty of time for Q&As.

Let us start with the headlines.

Sales in constant currency were up 4% to $4.7b.

Operating profits were down 3%.

Earnings per share were also down 3% to $0.54 per share.

In all cases well above the consensus estimates.

It is easy to conclude that the quarter was all about wholesaler stocking in the US, which boosted sales and the Sular gain from 2002, which depressed margins.

Undoubtedly these are factors.

I will say more on wholesaler stocking in a moment.

This would understate the pace of the transformation, which saw us able to absorb almost $700m of generic erosion in the US in the quarter.

This is the real message of the quarter.

Of course there are still a number of important milestones to reach this year before we can say the product transformation is complete.

Let us go through the P&L more carefully starting with sales.

I want to separate the sales performance into its two component parts, the underlying demand related to prescriptions and wholesaler purchasing behavior which is driven by speculation and the timing of future price changes.

As usual we have provided further details in the press release to help you complete your spreadsheets.

At the macro level, the business is performing more or less as we anticipated with generic erosion balanced by the strong demand for the portfolio of growth and recently launched products.

Both categories of which exceeded sales of $1b in the quarter.

In the US, pharmaceutical sales were up 3%.

IT is obviously after absorbing generic competition to Prilosec, Zestril and Nolvadex, where sales were down by 66% in aggregate but not in substantial decline of $669m.

But this generic erosion was more than offset by the growth in the rest of the portfolio, which grew by 55%.

An outcome also influenced by speculative wholesaler buying.

As I have said many times before, quantifying the impact of this is by no means a precise science.

Even after stripping wholesaler buying-out, we estimate the underlying demand for these products increased by an impressive 35%.

With the balance of growth or around 20% being attributable to wholesale stock movements in anticipation of price increases.

Price changes largely occurred on April 1st.

Our best estimate is at the end of the first quarter wholesaler inventories for our products were approximately $400m higher than normal.

It is also important to keep in mind that the first quarter of 2002 also included inventory building in the distribution channel estimated then at around $200m.

So it is stronger this year, but then of course the sales of these products are bigger too.

We believe that this excess inventory should work itself out in the normal course of business over the next two quarters.

We are not changing our estimates for revenue for the full year.

The products chiefly affected in the quarter were Nexium, Serquel, Toprol-XL and Atacand.

Again the press release has all the details.

Sales outside the US were up 5% with another strong performance coming from Japan where sales grew by 32%.

Before moving down the P&L, let me briefly review a few of the product highlights.

First Nexium.

It was another strong quarter with total sales of $835m, partly influenced by wholesaler purchases.

Our share of total prescriptions in the US continues to grow to 21.4% in March, up from 20.4% at the end of the year.

Our share in the GI specialist market has also increased and now stands at 27%.

The market is also up as well with PPI scripts up 14% in the year to March.

Overall our prescriptions for Nexium were up by 68%.

Turning to Prilosec, sales in the US were down 60% to $287m.

The decline broadly in line with prescription trends.

Our share of total omeprazole prescriptions has hovered around the 40% mark for a while and we are carefully monitoring three key variables, [indiscernible] supply capability which looks as though it may have increased recently; the overall demand for the omeprazole molecule; and our exposure to contracting wholesaler inventories, so as to ensure there are no surprises.

Iressa had sales of $19m in the quarter.

You will recall from last year that the initial uptake was very rapid, fuelled by pent-up demand at the point of approval.

It was always going to ease back in subsequent periods towards a more normal level of demand.

In addition to this easing back, sales have also been affected by labeling changes requiring closer medical supervision in Japan, which were implemented in response to case reports of [indiscernible] lung disease.

It currently looks as though the market in Japan is moving back towards a more balanced view of benefit and risk.

We will have to see what happens during the second quarter.

So far as the US is concerned, you all know that Monday is the [indiscernible] day.

Well there is nothing more to say at this point, other than we will all know the outcome in a few days.

Last amongst the products, and certainly by no means least, we have recorded the first sales for Crestor.

The first launch was in Canada on February 19th and was followed by the Netherlands and the United Kingdom in March.

Sales were of $300m in the quarter.

Although it is very early days, we are very pleased with the initial response in these markets.

In Canada for example, after 8 weeks, we have achieved a 23.5% share of new prescriptions in the private [indiscernible] market, which represents approximately 40% for the total market.

It is early days and too soon to reach conclusions for these individual markets and certainly too soon for any wider extrapolations.

I don't mean to gloss over the strong performances from some of the other products in the performance.

Seroquel, where market share continues to climb; good growth seen in Toprol-XL and Atacand;

Arimidex gaining momentum in the treatment of early breast cancer;

Symbicort with another good quarter.

But I am conscious of the time and they are all covered in the press release.

I will now complete the review of the financial items.

On foreign exchange, the weaker dollar added 5 percentage points to sales, bringing total reported sales growth to 9%.

Expenditures on R&D and SG&A combined grew by 4% in CER terms.

But on a reported basis, currency lifted the growth of these two cost items to 11%.

As a result, currency movements netted to a 1% benefit from the operating profit line and reduced margins by around 0.7%.

Currently we don't see the outcome for the rest of the year being very different from this.

In our terms, it is broadly neutral.

I have already mentioned margins and if you strip out the Sular gain from other income, operating margins in the first quarter of last year, margins were 26.8% in the quarter.

Virtually the same as this year's first quarter.

Product mix is improving gross margins to 76% in the quarter.

This benefit more than offset the small increases in R&D and SG&A as a percentage of sales.

Finally to our earnings guidance, which you will have seen is the same as we set out with the 2002 year-end results.

For EPS to be in the range of $1.50 to $1.65 per share.

We are clearly off to a good start with a strong top line and the cost base pretty well managed.

But obviously the wholesaler inventory will unwind and the additional profit generated in the first quarter will wash through in subsequent quarters.

We are of course striving to hit the upper end of this range, but many of the swing factors are still in play.

It is our view therefore that it is too early to narrow the range.

As far as the quarterly phasing from here is concerned, these first quarter results will clearly change the overall shape with the unwinding wholesaler stocking, taking place as we begin to prepare for the Crestor launch in the US.

Nevertheless, what is key is how the business exits 2003 with a new portfolio firmly in place and the momentum behind it as we go into 2004.

We will undoubtedly come back to questions on this, but for now we are ready to take your questions.

I now hand you back to the operator.

Thank you sir.

Ladies and gentlemen today's question and answer will be conducted electronically.

If you would like to ask a question, please press the star or asterix key followed by the digit '1' on your telephone keypad.

We will take questions in the order received and will take as many as time permits.

Again, ladies and gentlemen, please press star, '1' to ask a question.

We will take our first question today from Mr. Stewart Harris with UBS Warburg.

Please go ahead.

Yes, thank you very much.

Not wanting to dwell on the inventory bill, but nevertheless asking a question about it.

Just trying to come to terms with it.

I was just trying to work out a figure for the potential EPS impact of it.

Would I be a million miles wrong if I was to say $400m of stocking without any incremental SG&A, R&D costs, maybe a 60% EBIT margin, prevailing US tax rate come to something, high single-digit $0.08 or $0.09 benefit to the Q1 number.

That is the first question.

Secondly, just on Iressa, obviously you are not going to try and forecast what is going to happen.

Could you just let us know that under the approval process you are going for, what are the outcomes that the FDA could come up with?

Do we have approval or non-approvable or do we just have a straight approved or non-approvable?

That's it, thank you very much.

Okay.

You were very generous in giving me latitude of a million miles for the EPS impact of inventory build.

But I think in broad terms that is a sensible calculation.

If you take the expected gross margins with the fact that some of the products are in the Merck stream, then it pretty well drops through on an individual quarter.

Our Costs are completely independent of what goes on in the wholesalers.

I think it is of that sort of magnitude.

Bear in mind also that we are talking about, you are basing on the absolute level of stock.

As I have said many times, it is an imprecise calculation and we do have to make two key judgments.

One is what the normal level of inventories are and secondly what the underlying level of demand is.

I think the signaling will unwind over the rest of the year, or certainly much of it in the next two quarters.

It is not at all precise when it will come out.

I think on Iressa, it is virtually a binary outcome.

Whether it is approvable or non-approvable.

Okay.

So there is the latitude that the FDA could say approvable if, for example, they wanted to see more data.

It is not approved as in straight approved or --?

Sorry Stewart.

Let me just clarify that.

I think what we are pretty certain of is that we will get either an approval of the file or if the FDA so chose to do, a non-approval.

It is binary.

I think there is the outside chance that they may take some other route, but the expectation is it will be one outcome or the other.

We will know by the end of the 5th May.

Brilliant.

Thank you very much indeed.

Next question please.

From SG Cowen, we will take a question from Steve Scoular.

Please go ahead.

Thank you very much.

I have a question about Crestor's great success in Canada.

Maybe you could tell us why the Canadian private payer market would not be applicable to the US or other markets.

You have some appropriate cautionary commentary in the release.

But maybe you could tell us why you believe it is not applicable?

How is the market different?

Maybe you could give us 2 or 3 reasons why we should not extrapolate this huge success to the US?

Secondly, can you remind us of the price in Canada, the spectrum of usage and where that usage is coming from?

Yes Steve, I can take that.

Ed Seage here.

I think the primary caution on extrapolation is this is the early days of the launch and therefore you are dealing with relatively small prescription numbers being projected.

You have got wide [indiscernible] intervals.

So that certainly just a pure statistical caution there.

The private payer market as opposed to the broader market, obviously again you are talking about a sub-set of a national market that is smaller than the US market.

It is certainly way premature to take a straight line ruler, even to project it within Canada never mind take that ruler and then move it over to your forecasted line graph for the US market.

To remind you on the pricing, I think the pricing in Canada is about 15% below the prevailing prices of Lipotor (ph).

And where is the usage coming from?

What is the spectrum in terms of doses?

I think also on that Steve, I think we are not giving out any more data at this stage.

I think it is 8 weeks, a lot less than that in other markets.

It does look as thought it is clearly gravitating towards the 10mg as being the major dose, which is what you would expect.

I think it needs a while longer on the market before it really starts to level off.

I think by the time we get to the next report, we will have had 4 months or 4.5 months and we will see more clearly also what we expect the timing to be of the approval into the wider Canadian market.

This is not necessarily a quick process.

So I think in many ways it is best left to say, it is off to a pretty good start, but let us see a few more data points on it.

Okay, thank you.

Next question please.

We will now move to Mr. Andrew Baum with Morgan Stanley.

Please go ahead.

: Good morning.

A couple of questions if I may?

The first question, you mentioned Prilosec inventories.

Now obviously Nexium inventories are high and I wondered whether you could give us number of months for Prilosec inventories compared to what you have seen historically?

Second question, as much as you can, you mentioned the other European markets, whether there has been a similar kind of success in launch in those markets?

If you could update us there?

Thirdly, aside from stocking, in terms of maybe not your front-line products, you have had a fairly robust performance.

Do you have some sense of how you are going to maintain that performance when sales force tension is switched across to Crestor as Crestor gets rolled out in the various markets?

I will stop there.

Okay, thanks Andrew.

Prilosec inventory, it's a little like the first question, you have given me a fair bit of latitude in terms of number of months.

Well it is less than one and pretty well always has been.

Obviously it has peaked in the past where there has been significant speculation on it.

Clearly there is no merit whatsoever in speculating from a wholesaler's perspective on Prilosec.

So it is well below a month.

We have also been tracking it pretty carefully, both in terms of where it is in the chain, in the wholesale or the retail, what the likelihood of returns are and so on.

Really I guess for the last 4 or 5 quarters we have been managing the Prilosec inventory pretty carefully.

We certainly don't want or expect any surprises to come out of Prilosec.

The other European markets, well it is less than 8 weeks.

The UK market it is just a couple of weeks.

But I think we would make similar comments.

They both appear to be off to pretty good launches.

The data is even less mature than it is in Canada.

We are certainly pretty encouraged.

With regard to the performance of the other products.

I mean this again is something that we, as an organization, have been tackling openly for a very considerable period of time.

The launch of Crestor although it has been unpredictable, how we allocate the details particularly between Nexium and Crestor is something that we have been preparing with.

As you recall, when we restructured the sales forces in 2000, we combined the GI and cardiovascular forces for the sole reason for enabling them to do them together.

We are now constantly adjusting the distribution between primary and secondary details between the two of them.

Such that we believe that with the sales force that we have got, as well as the substantial productivity improvements that we have had from our US field force, that actually we can cope with both products pretty well.

Certainly it would be natural now for some of the detailing on Nexium to start to tone down a little bit as it has been on the market getting on for a couple of years.

I think we are confident we can cope with it and not diminish the prospects of either product.

It is what we are here for, to optimize the product performances within a given resource.

Andrew, I think we have said on a number of occasions that there is a very high overlap of the target doctors for both products.

So that obviously helps a lot in this situation.

Thanks Andrew.

Can we have the next question please?

From Merrill Lynch we will now turn to Mr. James Culverwell.

Good afternoon Jon and Mike.

Just a question on the product mix and the gross margin.

You comment that there was a favorable effect from Merck payments during the quarter.

Although I see that Losec sales were lower than Nexium sales during the quarter.

So how does that work?

Is that a legacy and how does it work for the next few quarters now in terms of that margin effect for payments to Merck?

Well I mean, obviously the big pieces of the payments to Merck relate to Prilosec and Nexium.

As a percentage of sales if Prilosec falls faster than Nexium, then the percentage of the payments to Merck goes down as a percentage of total sales.

So we are always going to see an improvement in the margin coming from that change in mix.

Indeed that is what we saw in the quarter, with about 1.2% improvement in gross margins.

Most of which came from the mix effect from Merck.

It wasn't the only improvement.

The mix will improve our underlying cogs (ph) as well.

But the major component of it was this Merck products taking a smaller percentage of the total sales.

There is obviously a little bit of a benefit in the calculation also because the Nexium rate is lower than Prilosec although offset by a higher earned manufacturing content.

That will also drive the percentage down slightly faster.

I am sort of assuming that the Nexium rate, I think ratchets it up a little bit doesn't it?

There must be some stage where Losec becomes de minimus and the Nexium therefore has a negative influence?

No, I wouldn't expect it to have a negative influence.

It is a lower payment to Merck and we are carrying more of the manufacturing stages on it.

We are seeing manufacturing, our ability to get some of the manufacturing costs down as the volumes grow.

So I would never expect Nexium to have a negative impact on gross margins.

If anything it is likely to be slightly positive.

I'm sorry but even if you have a significant payment to Merck coming from Nexium and Nexium grows as a contribution to the group, doesn't that mean that you have an increased contribution from a lower margin product effectively?

Mathematically yes, but if you look at the rest of the transformation products, Seroquel, Iressa when it comes, Arimidex, Casodex, they are not subject to Merck payments.

So the overall percentage of products associated with payments to Merck should decline.

That is where you get the relative -

Okay, yes.

I appreciate that.

Okay thank you.

Next question please.

We will now turn to Max Herman with ING Financial Markets.

Please go ahead.

Hi, good afternoon.

Just a quick question on margins again.

I mean obviously, I was very pleased with the progress in margins, but primarily on the stocking side.

I was wondering whether other than the royalty impacts from the Merck payments, there were any other reasons why the margin seems so strong particularly given the fall-off in sales of major products usually leads to a much more negative impact on the margins?

A second question just on Crestor.

Could you give us an update about the plans for that product in Germany and Spain?

I know that wasn't part of the original European approval?

Thank you.

Okay, thanks Max.

The question on margins, part of it goes back to the first question about wholesale inventory.

It doesn't drop through at the average margin, it drops through pretty much at the gross margin.

So you do get a disproportionate margin benefit, which is the same as what we saw last year, which is why the first quarter was ahead of the average.

That is an important part of it.

We have talked about the mix effect.

The third fact is that we remain absolutely determined to flatten off the rate of cost growth.

We have seen a 3% to 4% increase in R&D and SG&A.

Clearly SG&A at least is covering a higher level of activity.

There is a fair amount of Crestor work going on, on there now.

We are determined to keep it in the around 5% category that we talked about at the end of the year including the full launch of Crestor.

It is early days on that yet, but I think we are gaining more confidence on our ability to manage the cost line.

Not necessarily by slashing costs out, but by making sure that everything we do operates with a higher level of productivity.

Mike or Ed, anything you want to say about the European?

Yes.

The MR process Max, aims to get the common label across all of the countries.

As you know we agreed a common label with 12 countries and 3 dropped out, Germany, Spain and Norway.

The process from here on in is a process.

Each of their differences of opinion were different country to country.

So it comes back to a process of discussions with each of the individual countries and moving them towards a common labeling framework.

Any sort of timeline for launches in those markets?

I think it would inappropriate to give you a timeline.

Yes, when we have got a clear idea as to when the timeline will be, we will be in a position to say more.

It is relatively early days after withdrawal and we have got to go through the circumstances of each particular territory.

Thank you.

Next question please.

Next we will turn to Mr. Mark Becker with GIC.

Yes, hello Jon.

Just to probe in a little bit further on this gross margin question.

Looking at the Losec/Nexium proportion of sales in total, it has dropped from 36% to 33% year-on-year.

But it has dropped even further had there been no stocking, it would have dropped to about 30%, because it appears there was a disproportionate amount of Nexium stocking relative to the total stocking.

So would it be a mathematically true statement that your gross margin that you reported in this quarter is actually understated.

Because as we see this stock-out effect occurring over the next several quarters we will get even an additional boost on this mechanism that effects your gross margin?

Secondly, I am a bit befuddled by Mark's comment saying that the payments from you to them are going to increase at a double-digit rate.

Since the PPIs are a disproportionate amount of that revenue you pay to Merck, when in your own mind should we be modeling a flat revenue increase to Merck source from you through the various AstraZeneca/Merck contracts?

Unfortunately Julie [indiscernible] is not on the call here.

So it is going to be rather difficult answering the second part of your question.

I don't mean that in a facetious way because we don't know, I don't know precisely what assumptions go into Julie's calculation and I don't know what she starts at in order to be able to compute what the percentage growth is this year over last year.

We certainly give them advanced production forecasts, but we don't translate it certainly into [indiscernible] their books.

So it really is two quite different universes.

I think while we can help you with what we do with our own calculations, you have got to try and puzzle it out between us.

In terms of how it does impact us, it is pretty mathematical.

It is more or less taking the level of sales of the products effective and computing the royalties.

There clearly are some adjustments and some [indiscernible] that come up to that in subsequent periods, but it more or less follows that.

If the inventories were down in future periods, then clearly there is going to be a smaller proportion of Nexium sales in later periods that will come through in that overall mix calculation.

I will however caution you somewhat not to take cogs in a purely mathematical way, because I think as I have explained on some occasions before, you have really got to take account of the life cycle of a number of our products.

In some cases we are building capacity and certainly with Crestor we are building the throughput.

We are certainly nowhere near efficient production levels.

That means that we have got higher relative [indiscernible], there are other products that are at different parts of their life cycle.

So the totality of our gross margin doesn't necessarily shift in a linear fashion to the way the sales do.

The Merck piece of it is a bit more mathematical.

Thank you.

From CIBC, we will take our next question from Mara Goldstein.

Yes, I had two questions.

The first is just a follow-up on the Merck payment question.

Maybe so that we can better understand from a timing perspective, when you place orders with Merck for a product, what is the time delay that you then get delivery and expect to sell that through the market?

Secondarily, could you just talk about Toprol-XL?

I think there was a paragraph or a certification in the US and what the status of that is?

Yes, okay.

The production forecast like all major manufacturing relationships have a long-term, a mid-term and a short-term component.

You are able to adjust your long and mid-term forecast within tolerances.

Some of the supply chains have been very low, very short.

Prilosec has always been operated on a very short production timeline.

So it varies.

Generally, we look at fairly short-term forecasts.

But of course we have got to give the longer-term forecast to enable Merck to adjust their production capacity in line with demand.

We don't recognize the incremental payment or the sales related payment to Merck until we have sold the products.

So there are some intermediates accruing that we include in inventories for unrealized note payments.

Okay.

Overall, clearly we have got a paragraph 4 certification by JB Pharmaceutical (ph) against the 200mg dosage form of Toprol-XL.

So again, it is very, very early days in terms of those legal proceedings.

Just to remind you that Toprol-XL, the 200mg strength represents only about 12% of the total franchise in the US.

Okay.

You have filed a suit against them?

Yes.

Could you just remind us of the [indiscernible] date for Crestor in the US?

It will be about the middle of August.

Okay.

Thanks so much.

Alexandra Hauber with Bear Stearns has our next question.

Yes, good afternoon.

Sorry to have to dwell again on the inventory situation.

I apologize I missed one or two sentences of Jon's introductory remarks and some information has been given there.

Can you quantify roughly the excess stock levels or remind us of them in the last three quarters where you were pretty much running reserves $200m excess constant through the whole year?

I am particularly interested in your excess level at year-end?

Also, could you give us the IMS inventory levels for December and March for the five products that seem to be effected, Nexium, Toprol-XL, Seroquel, Atacand and Pulmicort?

I know they are not reliable, but it might give us an indication?

Can I just confirm, I think you said all these products have price increases in April?

Is that for all the five of them, or do any one of them not have a price increase in April?

I guess what I am after is particularly how much of those $400m excess inventories are actually related to Nexium and whether that has all been built-up in this quarter?

It is very difficult as I said earlier, it is an estimate and we do have to estimate normal levels of inventory.

That is one of the key components in therefore calculating what the excess is.

It is quite difficult when you have got a growing product.

If you have got a product, let us say Nexium for example, that may be started the year off with one month of inventory and ended up the year with one month of inventory, the dollar value will have gone in the chain will have increased quite substantially because the inventory in the chain protects future sales.

So you have got a number of moving variables.

I think the clearest way of looking at it, it pretty well all did work out through the course of 2002.

I think I quoted at the end of the year that we felt that inventory levels in total were pretty close to normal.

If anything I think I used the word "a touch below".

So I think for this purpose, I think we are reasonably comfortable in basing our analysis on the fact that there was no distortion at the end of the year.

Clearly product-by-product it may move a little bit.

But in aggregate we think that we entered the year with normal inventories.

We will have to track it each quarter to see how it unwinds.

I am pretty clear that it won't unwind in one quarter.

It may not all unwind in two quarters, but with the pattern of price increases the aim will be to exit the year as close to normal as possible.

So far as IMF data is concerned, it is not the only basis that we use for calculating inventories.

So even if I wanted to, I wouldn't be comfortable in giving you the data because it is just one mark.

Some of our products, we calculate inventory levels on tablets as opposed to prescriptions or prescriptions rates because you have got to convert prescriptions which vary from both 30 days to 90 days and so on.

We do use a variety of different mechanisms in order to do that.

I would not be comfortable in quoting you IMF data, because we don't use it in all cases.

With respect to price rises, we took price rises in the US on Nexium, Seroquel and Atacand at the end of the quarter, but not on Toprol-XL.

Okay.

I am still wondering whether you could give me some insight into how much excess in Nexium is there?

I also asked for the full year because I have the impression then based on my calculations that there was excess stocking of Nexium in the US and that was denied.

But if I try to add up the $400m where could I find stocking, that basically leaves me, I have about $200m or so excess stocking of Nexium in this quarter overall.

If that has all been build in this quarter that would be quite a lot?

Yes, well we said that three products essentially cover the excess inventory.

I guess a little over a quarter is Nexium, around a quarter is Seroquel.

The rest is Toprol-XL and Atacand.

You have got a whole series of small movements on things like Accolate and whatever.

They will add too much in any individual products.

What you've also got to remember Alexandra is that calculations can't be done off with net sales.

The calculations have to be done off gross sales, which is pre-discounts and rebates and return provisions.

They all have slightly different normalized levels of inventories where they are growing rapidly the dollar amount would go through the market.

Sometimes the prescriptions are lengthening, sometimes the dosages are increasing as they are with Seroquel.

I don't think you can do the calculations frankly with a high degree of confidence.

I think that may be a subject we can pursue on another day.

Can I have the next question please?

We will now take a question from Jason Culalis(ph) with Capital Guardian Trust.

Please go ahead.

Thank you.

Jon, I know you are reluctant to do any projecting off of the Crestor numbers in Canada.

It seems to me that you have had it now in the Netherlands for a number of weeks.

I am sure your marketing guys are straining your budget with IMS buys of data.

Could you tell us to what extent, if any, firstly the experience in the Netherlands supports the Canadian experience?

I think the pattern in the Netherlands looks similar to Canada, but obviously it is over a shorter period.

Now the analytical community has spent an awful lot of time agonizing over the side-effect profile of Crestor.

It would seem Canadian doctors have spent very little time?

Would you agree?

I think I would share your view that the early [indiscernible] Canada if you don't want to quantifiably extrapolate them suggest that the product profile and the message has traction.

Could you perhaps go into the message a little bit for us, just to remind us what it is?

Yes, I believe it is best in class, LVL(ph) lowering with the simple 10mg tablet starting dose, reaching your target cholesterol goals and the vast majority of the patients are going to start with.

It gets you right first time.

It is a very simple message and I think we have been clear about the safety profile as being very compatible with other [indiscernible] and we are getting thousands of more patient use now all the time to build that safety profile.

We are very comfortable with it.

The ethicacy method clearly has a lot of traction to it.

Now just finally, as you approach the August [indiscernible] date or whatever it is, the FDA is a little bit risk averse if you will and they might like to take some outside advice just to sort of CYA (ph).

Do you think there might be another advisory committee?

Well anything is possible and the FDA will publish its agenda ahead of time.

We will all be aware then.

We will be very happy to take the opportunity to present the full benefit and safety profile of Crestor whatever questions are asked.

Thank you.

We will take our final question today from Mr. Mark Clark with Deutsche Bank.

Yes, good afternoon gentlemen.

I am just slightly perplexed by the Iressa number.

I hear what you say about the pent-up demand and the label changes.

But I think you mentioned sort of well over 20,000 patients being treated and we know what the price of the drug is.

So it tends to imply that there has been a significant reduction in patients on the drug unless I have missed something.

Firstly, could you talk through that?

Secondly, could you tell us how many patients roughly are currently receiving treatment in Japan?

Well I don't know the answer to the second question.

We will flick through the pages and see if we've got it.

I think there is no doubt that the last quarter of 2002 reflected a higher than normal level of trajectory.

It was reflecting a large number of patients coming on and it was not normal.

We would have expected that to normalize during the first quarters of this year.

What we actually saw was instead of normalizing it, it went significantly the other way.

Undoubtedly the discussion on ILD in Japan has influenced both patients and physicians.

The [indiscernible] of supervision of prescribing did reduce the number of new patients going on to it and our view is that it has taken it now well below normal.

I think we will see over the next couple of months that it should start to track back up again.

There are obviously are a lot of complex factors in play, not just medical, but psychological as well we have had to address all of those.

Do we have a number or not?

No, we don't I'm afraid.

Okay.

Can I just ascertain one final point?

We had a call with a clinician who does a lot of work with the [indiscernible] and I think he's got 100 patients on it in the compassionate use program.

According to him you are set to close the compassionate use program irrespective of the FDA decision next week, so approval or non-approval, it shuts down.

Is that correct?

Let me explain the situation.

If we get approval, the compassionate use program, the EAP, the Expanded Access Program, which is the pre-marketing program will close and patients will either go onto the commercial product if they have reimbursement.

If they don't then we have the post-marketing compassionate use program, the patient assistance program that we have done for other drugs and Iressa will become part of that program.

So we will move to the normal operational position that we have with other cancer therapies and other therapies in the US.

Should we get a non-approval, then we have to consider the future of the Expanded Access Program in the context of what clinical programs will be required to move the drug forward.

That debate will take place when we know the conditions of the non-approval.

Hopefully, we won't have to answer that question.

Okay, I'm sorry to [indiscernible], but very briefly you also suggested there might be reimbursement issues being an oral drug.

Do you anticipate any reimbursement issues in the States.

Well it is clear it won't be reimbursed under Medicare, or cancer therapies would not be reimbursed under Medicare, but one would expect that private healthcare will reimburse the product.

Thank you.

That will conclude today's question and answer session.

Ladies and gentlemen, I would now like to turn back to Mr. Symonds to any further remarks.

Well thank you all for your time and attention.

I know this is a particularly busy day for you.

I am sorry to add to your burden, but thank you all very much for joining this call.

If you have any further questions, then you know how to contact our Investor Relations Department.

Thank you very much and good afternoon.

That will conclude today's conference call.

Thank you for your participation ladies and gentlemen.