Myriad Genetics Inc (MYGN) 2005 Q3 法說會逐字稿

Good morning. My name is Patrice, and I will be your conference facilitator. At this time I would like to welcome everyone to the Myriad Genetics third-quarter financial results conference call. (OPERATOR INSTRUCTIONS) Mr. Meldrum, you may begin your conference.

Thank you. Good morning and welcome to the Myriad Genetics conference call for our third fiscal quarter, which ended on March 31, 2005. My name is Peter Meldrum, and I am the President and Chief Executive Officer. I am joined here today by Jay Moyes, our Chief Financial Officer, and Adrian Hobden, President of Myriad Pharmaceuticals. I will begin the discussion this morning with a brief review of the past quarter, and will be followed by Mr. Moyes, who will discuss our financial results. Dr. Hobden will then discuss the Phase I cancer trials that are currently underway and will address questions concerning the recently announced preliminary results of our Phase II study of Flurizan for the treatment of Alzheimer's disease. At the end of his presentation I will turn the call back over to the operator for the question-and-answer period.

Please note that some of the information presented here today may contain projections or other forward-looking statements regarding future events or the future financial performance of the Company. These statements are based on management's current expectations, and the actual events or results may differ materially from those expectations. We refer you to the documents the Company files from time to time with the Securities and Exchange Commission, specifically the Company's Annual Report on Form 10-K and its Quarterly Reports on Form 10-Q. These documents identify important risk factors that could cause the actual results to differ materially from those contained in our projections or forward-looking statements.

The Company again enjoyed a strong revenue growth this past quarter with its predictive medicine business. For the third fiscal quarter ending March 31, 2005 predictive medicine revenues increased to a record 18.4 million, as compared to 11.7 million for the same quarter the prior year, an impressive 57% growth in revenues. Myriad's increased sales and marketing efforts, supported by journal articles demonstrating the clinical utility of our products, have resulted in the wider acceptance of our products by the medical community.

To make our predictive medicine testing available to more individuals we have been working with the professional nursing organizations, particularly ONF and OES. Myriad has developed educational programs to assist advanced practice nurses to incorporate genetic counseling into their clinical practices. This expands the line of services that they offer in their practices and makes our predictive medicine testing more available to their patients. Several hundred nurses have been trained through these efforts, and their clinics have now become new customers for our predictive medicine products. We anticipate continuing these training efforts into next fiscal year.

We have continued to work with insurers to streamline the reimbursement process for our products and to make it easier for patients to qualify for testing. Changes in insurance coverage guidelines have expanded the number of individuals who are now considered appropriate for testing. Several large insurers have recently recognized the value of eliminating the cumbersome pre-authorization requirement for our predictive medicine products. We believe that both expanded coverage in the individuals and the elimination of pre-authorization will lead to an increased number of tests.

Continued technology improvements and efficiency gains in the operations of our predictive medicine business contributed to a 71% gross profit margin for the three-month period ending March 31, 2005, as compared to 68% gross profit margin for the same quarter in the prior year.

You may recall that the royalty obligation for our PCR amplification license concluded at the end of March with the expiration of the PCR patents. The cost set factor savings will positively affect the Company's gross profit margins in the future, as Jay will discuss later on in the conference call.

The net loss for the third quarter of fiscal 2005 was reduced to $10 million, or $0.32 a share, compared with 10.7 million, or $0.39 a share, for the same quarter of fiscal 2004. The small decreased loss reflects gains in our profitable predictive medicine business, despite the Company's increased investment in drug development with five human clinical studies ongoing.

All of us at Myriad are very excited about our future with first in class compounds in the clinic that address large unmet patient needs, a strong preclinical pipeline that is rapidly moving toward clinical testing and a profitable and growing predictive medicine business.

I would now like introduce our CFO, Jay Moyes, who will provide more detail on the financial analysis.

Thank you Pete. It certainly is a pleasure to present a more detailed look at Myriad's financial results for our third fiscal quarter ending March 31, 2005.

As Pete mentioned, we are very pleased to report that predictive medicine revenues this quarter have once again hit a record. Predictive medicine revenues for the quarter ended March 31, 2005 were $18.4 million. This result represents a 57% increase over the same quarter in the prior year. The predictive medicine revenues for the nine months ended March 31, 2005 are up 67% over the same nine-month period of the prior year. The solid 18.4 million performance this quarter is in line with the analyst consensus predictive medicine revenue forecast. We have observed that sample flows continued to be strong and we expect that our predictive medicine revenues will increase again in the fourth fiscal quarter.

Our gross profit margin on predictive medicine sales this quarter was 71%, which is a 3% improvement over the 68% margin in the same quarter ending March 31, 2004. Because the PCR patent expired at the end of March, our cost of sales should decrease by approximately $1 million in our fourth fiscal quarter, which should enable our gross margins to improve to the mid-70% range.

The number of days sales outstanding continues to decline. The average days sales outstanding for the quarter was 79 days, as compared to 85 days last quarter and 96 days for the quarter ending June 30, 2004. This substantial improvement is a direct result of information technology enhancements combined with the dedication of our internal billing and collections team. The quality of our accounts receivable continues to improve.

Research and development expenses for the quarter ending March 31, 2005 were $15.5 million. This represents a 25% increase over the same quarter in the prior year, and an increase of 7% over the prior quarter. This increase was primarily due to costs associated with our human clinical trials in prostate cancer and Alzheimer's disease, as well as the enrollment of three Phase I trials for Myriad's two new cancer drug candidates MPC-2130 and MPC-6827. We have also made significant expenditures in the late stage preclinical development of our other drug candidates. Since we have five human clinical trials under way we expect our research and development expenses to grow modestly over the next several quarters.

Selling, general and administrative expenses for the quarter ending March 31, 2005 were $9.8 million compared to $8.8 million in the same quarter of the prior year. This modest 11% increase over the prior year was largely attributable to increased sales and marketing expenses, commissions and expenses incurred to support the 57% growth in our predictive medicine revenues. We expect our selling, general and administrative expenses will continue to increase depending on a variety of factors, including the number and scope of new product launches, our drug discovery and drug development efforts and our growth in predictive medicine revenues.

Our net loss for the quarter was $10 million, or $0.32 per share, compared with $10.7 million, or $0.39 per share, for the quarter ending March 31, 2004. We were pleased that this result was better than the research analyst consensus loss of $0.33 per share. This modest net loss is the direct result of a significant earnings contribution made by our predictive medicine business, which was $5.1 million before taxes, depreciation and amortization for the quarter ended March 31, 2005. Additionally, the operating margin of our predictive medicine business for the third quarter was 25%, which represents a significant improvement over the 16% operating margin generated for the same quarter of the prior year. We believe that our revenue growth will enable the Company to maintain a relatively stable net loss through the end of the fiscal year, even though costs associated with our clinical development programs will be increasing.

Cash, cash equivalents and marketable securities was $115.3 million at March 31, 2005. This compares to $124.7 million for the period ending December 31, 2004. We again point out that Myriad has no debt and no convertible securities, and that the total number of shares outstanding at December 31, 2004 -- excuse me, the total number of shares outstanding at the end of this quarter was a very modest 30.8 million shares.

Thank you for your attention. I will now turn the conference call over to Dr. Adrian Hobden.

Thank you Jay and good morning. This morning I am going to provide an update on our cancer compounds in clinical development and to address some questions that arose in Monday's conference call regarding the results from the Flurizan Phase II trial. Firstly, I will discuss our two cancer compounds.

As you will remember, we submitted INDs to the FDA in December and received permission to proceed with three Phase I clinical trials for our cancer drugs MPC-6827 and MPC-2130. In addition, we recently reported preclinical data on MPC-6827 and MPC-2130 at the American Academy of Cancer Research meeting.

Enrollment is proceeding on schedule in each of the three Phase I clinical studies for our oncology drug candidates, MPC-6827 and MPC-2130. You may recall that neither of our cancer drug candidates is susceptible to multiple drug resistance and both have demonstrated excellent activity in animal models.

MPC-6827 in particular has exciting potential for the treatment of brain metastases due to its remarkable ability to cross the blood-brain barrier. In animal models, MPC-6827 had about a 15 fold higher concentration in the brain than in plasma, yet did not appear to accumulate because the half-life of clearance of the compound from the brain was about the same as clearance from plasma.

Metastatic brain cancer is one of the greatest challenges facing oncologists today with over 180,000 new cases of brain metastases reported each year in the United States. Treatment options for such patients are very limited. There are currently no cancer drugs approved for the treatment of brain metastases because most oncology drugs will not cross the blood-brain barrier. Unfortunately, patients often die from the effects of their brain metastases, even though their peripheral disease is under control. It is for this reason that our investigator at MD Anderson has elevated MPC-6827 to the highest priority for testing in patients.

Interestingly, MPC-2130 also has good C&S penetration, although not at the elevated levels seen for MPC-6827. MPC-2130 also has good activity against blood cancers and is currently being explored not only for activity against peripheral cancers, but also for leukemias that have spread to the brain. Trials are proceeding at the MD Anderson Cancer Center in Houston, Memorial Sloane-Kettering Cancer Center in New York, and the Huntsman Cancer Center here in Salt Lake City.

I would now like to follow up on questions from Monday's conference call on the Phase II Alzheimer's trial, and reiterate some of what I said at that time.

As required by the FDA, Myriad submitted a statistical analysis plan for our Phase II Flurizan trial prior to the efficacy data being unblinded. In that plan we identified the primary and secondary analysis. Our analyses use the intent to treat population, which is required by the FDA. Results from these analyses are the primary data that are submitted to the FDA for regulatory consideration. Any other analyses are primarily for internal use.

Our primary comparison, as stated in the protocol, was 800 mg BID dose against placebo -- that is, all patients on 800 mg BID versus all patients on placebo. The analysis showed that the 800 mg BID patient group consistently performed better than the placebo patients on ADAS-cog, ADCS-ADL and CDR Sum of Boxes. However, the results were not statistically significant. This result may be accounted for by the surprisingly low rate of decline of the placebo patients, which, as I reported on Monday, was 4.1 units on ADAS-cog when based on previous trials 6.3 units was expected.

Our secondary analysis was a comparison of 800 mg BID against placebo for the two separate groups. Mild patients, as defined by an MMSE of 20 and above, and moderate patients, as defined by an MMSE of 15 to 19. Prior to unblinding the study, and in accordance with FDA guidelines, we excluded patients from two sites from the intent to treat analysis because of procedural violations. Therefore, our total number of patients was 189, of which 68%, or 128 patients, were defined as mild. The data that we reported on Monday in mild patients was therefore a pre-specified analysis and not a post-op analysis of the data.

As we reported Monday, in mild patients there was a strong trend towards activity for the 800 mg BID group against the placebo group. Indeed, if we had used a one-sided test rather than a two-sided test, we would have seen statistical significance for both ADCS-ADL and CDR Sum of Boxes.

The 400 mg BID group did not reach statistical significance even by this test, but did show a positive trend.

Since most patients in the trial had mild Alzheimer's disease, there were few patients in the moderate group and only 15 of these moderate Alzheimer's patients were randomized to the placebo group. Additionally, a large amount of variability occurred in the moderate group, as is typical for Alzheimer's patients with fairly advanced disease.

Based on our preliminary analysis, the data shows that the drug treated groups and placebo group were essentially equivalent. The test scores in drug treated group were sometimes slightly above placebo and sometimes slightly below. There was no discernible trend in either direction. Therefore, our conclusion from the trial is that there is strong evidence that Flurizan is working to slow the progression of the disease in mild Alzheimer's patients, but there is no evidence for activity in more advanced patients. This result, we believe, is consistent with the amyloid hypothesis of Alzheimer's disease, and consistent with the results from animal studies on Flurizan.

Thank you for your attention. I will now hand the call back to Pete.

Thank you. I'll turn back to the operator for the question-and-answer portion of the call.

(OPERATOR INSTRUCTIONS) Charles Duncan, JMP Securities.

Thank you for taking my question and congratulations on a strong quarter. A couple of questions with regard to predictive medicine fundamentals. You talked about the growth in revenue. Can you tell us about the drivers? Was there a price increase impact versus demand trends, Jay?

Charles, I think the impact of the price increase, which was last April, so a year ago April, is pretty much already filtered through the sales for the past -- I really don't think it had much effect this quarter.

Charles, this is Pete. I would agree with Jay. If you look at the sample flow, it tracks percentage increase wise very well with the revenue growth. So I think we're seeing all increased volume and demand. There's a very little effect in terms of the price increase of a year ago.

As I mentioned also on the call, Charles, we're seeing good trends in the next quarter and seeing the sample flow continue to grow.

And then with regard to some of the changes in the insurance guidelines that you mentioned, Pete, with regard to pre-authorization, as well as expansion of coverage, could you provide us a little bit more color on some of specifically what some of those changes have been?

Some of the larger insurance companies in the past have required pre-authorization, and that does become cumbersome for the patient. The patient will usually see a physician, have a blood draw, have that sample shipped to Myriad, but then we'll have to wait until the insurance company authorizes prior to our doing the test the performance of that test. During the past quarter several of our larger insurance companies have seen the wisdom of doing away with that pre-authorization as we see predictive medicine tests becoming more standard of care in the treatment of hereditary cancers. And this enables the patient then to immediately to have that sample analyzed and the results provided back to the patient without any delays waiting for insurance approval. And this required a clarification then of the criteria for testing patients, and on the part of the insurance companies, a doing away of any pre-authorization requirement.

So that clarification of criteria has also expanded the covered -- the number of patients that could use the test?

Yes. The criteria has evolved over the past several years, and initially that criteria is relatively restrictive. And as we learn more about the value of the test and the use in terms of managing the patients' health care, we have seen those criteria relaxed by the insurance company and the test being made more available to patients with a broader criteria.

And then with regard to data that may be used to continue to drive broader use of tests, are there any ongoing studies that you know of that you can point to that you look forward to the results of? Or do you think that currently published data is sufficient to continue to drive demand both for BRACA, as well as the COLARIS (inaudible)?

There have been a number of articles recently that demonstrate the clinical utility of our test. For example, a number of studies have been published showing that MRI is much better than mammography in young women in terms of identifying cancer, identifying tumors. And there are additional studies under way to confirm that.

We have also seen a number of studies showing that for older women who have had cancer 20 or 30 years ago and the cancer has come back and begun to spread, that this test can be very useful in terms of selecting the most appropriate chemotherapy for treating their metastatic disease.

Additionally, there are tests under way looking at BRCA2 and its role in pancreatic cancer, and looking at these tests again for additional therapeutic benefit.

So we've been fortunate to have a number of very positive articles demonstrating the value of these tests. And there are several clinical studies under way that will be confirming that data.

My final question is for Adrian. Flurizan, in our review of the literature for all of the other approved drugs it's clear that most of them look at mild to moderate. They also do some of these other analyses like, for example, focusing on ADL, etc. But in addition, we've seen approximately only a 50% compliance rate for those other drugs. Can you tell us about the experience in your trial? Do you think that you have approximately 50% compliance? Maybe was it better than that? Does it differ between mild to moderate, etc.?

I think as I said on Monday, Charles, that since we don't have a tablet count lot yet we can't be specific about -- we can't be for sure (ph) about how compliant the patients were with the entire protocol. We think that there was about 80% compliance by other criteria, but that's only by other criteria. I think general experience is that compliance is usually around about 60% when you look across most trials. So if we did get 80% that would be better than the average. But it's kind of too early to say at this moment.

Could you clarify the statement that you made regarding the moderate patients and the number randomized? It sounds like the variability in that was maybe greater than the mild variability.

If you do the math and realize that we had 189, of which 128 were in the mild group, then that automatically makes it a small group for the moderate patients. And we didn't stratify for across the groups. So we actually had far fewer patients in the placebo group than in the drug groups, particularly in the 400 mg group.

But you just see a lot of variability in the various outcome measures in moderate patients and a good deal of scatter in there. And so it's taking some time to really figure out what's happening there. But as I said in my -- earlier on, we see the patients, the treated group patients, sometimes they're doing better than placebo and sometimes they're doing worse through the 12 months, but only by a small amount. And so we just don't see any trend to benefit or to detriment in those patients.

Finally, clarify your statement regarding the one tail versus (multiple speakers)

Just a statistical analysis. A two-tailed test is what the FDA requires. It requires you to test the chance that your drug either is a benefit to the patient or made the patient worse. A one-tail test would simply test the hypothesis that it made the patient better, and so it's not acceptable to the FDA, but it's very acceptable in a scientific level.

Okay, thanks.

Annabel Samimy, UBS.

I have several questions related to both predictive medicine and Flurizan. Can you give us an idea of the sample flow growth? Is that a -- the consistency that you're seeing, is that a year-over-year number? Is that a quarter-over-quarter number? And -- well, that's the first question. I will leave it at that and I will continue to ask more.

Yes, we have seen steady growth in the samples on a year-to-year basis and over a quarter-to-quarter basis. So the demand for the predictive medicine test for a number of the factors that I alluded to has continued to grow on a fairly consistent rate from quarter to quarter and certainly were up very strongly year to year.

And that's consistent with actual revenue growth, the sales growth?

Yes. As we mentioned earlier, this is being driven entirely by demand and not price increases. It is sample flow growth.

With that demand are you seeing an expansion in the actual population sites that are taking the test? Or is it still generally in the same population that you have been targeting before, your general oncology patient?

No, we are seeing expansion in two areas. As you have alluded to, our primary sales force focuses on oncologists and newly diagnosed patients with cancer. But we have expanded into the primary care market, focusing on OB/GYNs and young women who have a family history of cancer that do not yet have the disease. And we have clearly seen an uptake in that segment of the market.

Also, as I mentioned, because of the value of these tests in guiding the decision with regards to chemotherapy for treatment of metastatic disease, we certainly have seen an increase in older women who have been diagnosed with cancer a number of years ago for specifically the treatment of the recurrence of the disease in individuals. In fact, it is sometimes surprising for individuals our second largest customer is Medicare. And obviously it's our older women who have seen the cancer recur.

So we are seeing clearly an expansion on both ends of the spectrum for women and to older women with metastatic disease and younger women that are pre-symptomatic.

In terms of backlog for genetic counseling, to what extent are you still seeing -- I know you've expanded the number of customers who can use counseling and then use your test. But do you still see a general backlog in genetic counseling? Also on that note, what exactly did the timing of the pre-authorization restriction lift? And was there any kind of backlog with -- what kind of backlog did you have related to that during the quarter?

Thank you. We had two major insurance companies lift the pre-authorization restrictions this quarter, and both of those occurred more toward the end of the quarter. But the Company -- I'm sorry what was the first part of your question?

It was the backlog related to those restrictions, like what is the lag that you see in terms of the authorization (multiple speakers)

It is hard to quantify any lag or backlog associated with patients waiting for genetic counseling. We do believe that's an important dynamic in this business, and that's why we've been focusing on the training of nurse oncologists to handle that genetic counseling function. As I mentioned, so far this year we've trained several hundred. There are probably only 300 or so genetic counselors trained in oncology. So we have essentially this year doubled the number of genetic counselors to handle patients. And that has to have a significant positive effect in terms of reducing the backlog. But it would be hard for me to quantify that more specifically.

Let me move on to Flurizan really quickly. Could you clarify what you were talking about in terms of your primary comparison and your secondary comparison? You said it was 800 mg versus placebo, and then the secondary analysis 800 mg BID in the mild and then moderate -- can you just clarify that? I didn't quite catch that.

The FDA, and actually other regulatory authorities, require that you put in a statistical analysis plan before the trial is unblinded so that you are kept honest, if you will. You say exactly what you're going to do and what you're not going to do, and what you're going to report to them. Because if you do too many different types of analysis then it gets rid of any statistical significance. So you have to limit the number of analyses that you do from a regulatory perspective.

So as required by FDA, we prespecified to them what our primary analysis would be and then after we had done that analysis, what our secondary analysis would be. So our primary analysis we specified was to take the high base dose group of patients. We reasoned that the high dose group would likely respond better than the low dose group. So 800 mg BID, and compare that in the whole group of all patients eligible for analysis versus the placebo group all patients eligible for analysis. That was our primary analysis.

And that was where you saw the 44, 41 and (multiple speakers)

No, that was in our secondary analysis of the mild patients. So in our patients overall, which includes both mild and moderate, we saw 800 group doing better than placebo, but we did not meet statistical significance. So when we did our secondary analysis, which was to take out from the total population the 800 mg, so the mild patients, and looked at 800 mg BID mild patients versus placebo mild patients, that's when we saw the statistical significance by a one-sided test. Have I clarified that for you?

Yes, thank you. I'll get back in the queue. Thanks.

David Webber, First Albany.

Thanks for taking the question. First question for you, Pete. Again getting back to some of the changes that the insurance companies have made, can you give us any detail on exactly how the criteria for eligibility have been broadened in terms of family history, or age, or other factors?

It varies somewhat by insurer and insurance company to insurance company. So it's hard to be too specific other than to say that when we initially launched these tests the criteria tends to be more restrictive because the tests are new and physicians have less experience with them. And we have noticed over time as we begin to demonstrate the clinical utility and the value of the test that those guidelines tend to be less restrictive. But it really varies from insurance company to insurance company, so it's hard for me to be much more specific than that.

A question for you, Adrian. I don't know whether you have opined about this yet, but the reason that you think might explain the slower than expected cognitive decline in the placebo patients?

I wish I had a reason, David. Actually, if you break it down by mild and moderate, the mild patients declined on placebo just slightly slower than you would have expected historically, but our placebo group for the moderate patients declined a lot less than we would have expected; actually about half of what we would have expected. And so we don't know whether there is a general slowing -- a general decrease in the rate of which placebo patients are declining in the population as a whole -- and to be honest, that has been suggested by various experts in the area -- or whether we actually have just an aberrant moderate Alzheimer's group in the placebo group. And that is a distinct possibility, because I think a halving of the rate of decline is something which is not to be expected.

The trouble is with small trials like ours, especially when you only have 15 patients in the moderate placebo group that can have a huge impact on the results of your trial. So I think it's going to be very interesting to look at this in a much larger number of patients where an individual patient can't have so much influence.

Thank you.

Ladies and gentlemen, we have reached the end of the allotted time for questions and answers. Are there any closing remarks?

No, thank you. But I would like to indicate to everyone on the conference call we appreciate your continued interest in Myriad and thank you for listening in this morning.

This concludes today's conference call. You may now disconnect.