Eisai Co Ltd

TSE:4523

Earnings Call Transcript 2019-Q2

[Unknown Executive]

Thank you very much for taking your time to attend our financial results presentation. We would like to begin the financial results presentation session for the second quarter of fiscal 2018. Please make sure that you have all of the materials in front of you. Please find the presentation slide and last report and reference materials. If you find any of the document missing, please raise your hand.

I would like to introduce the attendee from the management. Director Representative and CEO -- Corporate Officer and CEO, Mr. Haruo Naito.

Without further ado, I will give the floor to Mr. Naito.

[Haruo Naito, executive]

Now I would like to start briefing on the financial results for the first half of FY 2018 and current state of our business. You will see the actual results on this page.

As one of the main topic for today is that through partnership model, we were able to achieve increasing revenue and profit. At the same time, we made a proactive investment in R&D.

Revenue was JPY 310.1 billion, which was up 9% year-on-year.

Growth in global [ full ] brands, particularly in LENVIMA, Fycompa, grew significantly.

By region, Japan, Asia, Latin America and China, in these regions, we grew. And together with Merck, HCC approvals brought about the milestone payments from them. We received the milestone payment and, therefore, revenue could grow up 9% year-on-year.

Cost of sales improved by 6.1 percentage points. Of course, key factors for this was -- although there were the -- already the new drug price in Japan, although it is the factor for increasing the cost of sales, but in -- for Fycompa and LENVIMA, cost structure for these agents is very good, so growing these sales.

As for ALOXI, actually generate -- eroded the market for ALOXI, but the cost was pretty high for ALOXI. Therefore, the product mix in terms of the improvement of the cost of sales has been reduced. Of course, the onetime payment received from Merck also contributed to decreasing the cost of sales. As a result, gross profit reached to JPY 218 billion, up 19% year-on-year.

R&D expenses were JPY 65 billion, accounting for 21% of the sales, down from 1 year earlier, which I would like to explain later.

In our case, we have a partnership model. Our partners, Merck, Biogen, Purdue, we asked them to share cost of R&D activities, so they bear such burdens. And including such sharing, R&D expenses grew over 20% year-on-year and made a such proactive investment.

SG&A expenses grew 17% from 1 year earlier, which included most of the sharing of profit of JPY 7.9 billion with Merck. And currently, LENVIMA is being launched for HCC globally. For this, we have made the proactive expenditures of the budget.

And anyway, overall, expenses growth was controlled within the growth of the gross profit. Therefore, we have kept our financial discipline. As a result, operating profit was JPY 48.4 billion, which was 74% increase from a year earlier. Significant growth in profit was achieved. Similarly, profit for the period and profit attributable to owners of the parent, both grew over 70% year-on-year.

ROE 10.7% at the interim result, with 10.7% for both -- the capital cost is 8%, then we have a positive equity spread of 2.7%, bringing about the increase in the shareholders' value. For your information, double-digit ROE has been achieved first time since the first half of FY 2013.

Free cash flow was generated in the amount of JPY 45.4 billion. This amount exceeds the total amount of annual dividends of JPY 150 per share. So that means that we have been able to generate the cash covering or over the total amount of annual dividends during the first half. At the Board of Directors meeting, there has been the resolution to pay a JPY 70 per share as interim dividend. And the net DER was improved to 0.31 -- minus 0.31, and equity ratio is 58%. Therefore, robust mix and healthiness of the financial structure has been strengthened further.

Here is the breakdown of the revenue migration with this waterfall chart. First, JPY 8.4 billion has been added by expansion of global brands; JPY 6.8 billion from Japan business, which includes the onetime payment received by us for return of a marketing right for Lipacreon. China and Asia, which are growing markets, contributed JPY 5.8 billion. Milestone payment for LENVIMA's HCC indication approval contributed JPY 22.2 billion. In Japan, in U.S. and China and Europe, since the beginning of this fiscal year, in the U.S., Europe and China, we obtained this milestone payment, and the total amount was JPY 22.2 billion. There was a decrease in ALOXI revenue because of the erosion by generics in the market. And as a whole, revenue increased by JPY 25.1 billion to reach JPY 310.1 billion.

Similarly, if we look into the breakdown operating profit migration, similar items are listed here. Global brands have contributed JPY 7.5 billion, irrespective of the impacts of the NHI drug price revision. JPY 4.4 billion increase was seen in Japan business. And the China and Asia business, JPY 4.4 billion and JPY 22.2 billion was contributed by the milestone payment for LENVIMA. And the decrease in ALOXI profit of JPY 7.7 billion, that was some onetime event. And the JPY 7.9 billion cost of a profit sharing with Merck is included in this JPY 10 billion. And offsetting all of these, operating profit increased by JPY 20.6 billion, reaching JPY 48.4 billion.

On your right-hand side, you see the breakdown of R&D expenses. Earlier, on the P&L, JPY 65 billion were to spend, but adding JPY 24.3 billion, as I said earlier, the cost of R&D is [ worn ] by partners. We received the reimbursement from them. And adding them back, actual expenditure in R&D activities was JPY 89.3 billion, accounting for 29% of sales. Compared to 1 year earlier, this is up 20% from 1 year earlier. Actual expenditure in R&D increased by 20%.

So through partnership model, we have been able to allocate more -- significantly more resources into R&D expenses. This amount, JPY 89.3 billion, needless to say, is spent for BAN2401, elenbecestat and aducanumab next-generation candidate in AD. Lemborexant, LENVIMA, very rich pipeline candidates. We have been able to spend this amount into these candidates.

Regarding the R&D for LENVIMA, the cost sharing with Merck as well as the factual -- the substantive R&D expenses was 0 because of the reimbursement from Merck as well.

Profits for first half. In-house global brands were the growth driver. The revenue of 4 global brands was JPY 56.2 billion, up 27% year-on-year. For LENVIMA, as well as for Halaven, we have seen growth. Fycompa grew over 30% year-on-year.

LENVIMA is approved for HCC globally. And all over the world, we have started full commercialization with Merck. Fycompa was approved in the U.S. for pediatric patients. And we submitted for the partial-onset seizure in China. In Japan, as well, we aim to submit for monotherapy in Japan in Q4. BELVIQ data from long-term cardiovascular outcomes trial, which is obligation for post-marketing periods, and this trail has been completed. It is a large-scale trial with over 18,000 patients, type 2 diabetes. We have obtained a new data regarding the prevention and remission of type 2 diabetes. Positive data has been published in The Lancet.

Next week, we are going to have opening ceremony for new Suzhou Plant in China, of about 40,000 [indiscernible] as area and reach the capacity per year is approximate 3 billion tablets in formulation and 5 billion tablets in packaging, almost a -- commensurate to the size of the Kawashima Plant capacity. And this will become fully operational first.

Now I would like to give you the report on dementia, mainly from the presentation at CTAD. I'm going to cover these 4 studies. First, BAN2401 protocol design. We have used adaptive designs for the protocol. You may think that you are not familiar with this design. You may have had various questions about this design. Please look at this document. It said Adaptive Design for Clinical Trials of Drugs and Biologics, Guidance for Industry by FDA. FDA published this guidance for adaptive designs in September this year. BAN2401 protocol design was already completed and made years before this draft guideline was issued. But protocol design of study 201 -- or study for BAN2401 satisfied the key principles or guidelines. So this design, it can be said that it is a legitimate design. And Bayesian approach is a legitimate statistical method and enables the early decision-making and early evaluation for treatment effect by randomized -- randomizing more patients to treatment groups with higher response based on interim analysis. It is unconceivable if you are based upon fixed design. We believe that there is a big advantage. Frequently, IA, interim analysis, is repeated.

So regarding the results of the interim analysis predetermined are controlled in terms of access. I myself have no opportunity to know the IA results. By repeating the IA repeatedly, Type I error is controlled appropriately. Data integrity and the reproducibility can be assured securely in this study.

Predefined set of rules. At the end of the double-designed treatment period of 18 months, conventional statistical analysis was used to assess the treatment effect side. That was the predetermined rule. Key value study was test -- was conducted in order to assess the magnitude of treatment effect.

Next, as for ADCOMS, you may think that this is something that we coined, twisted, so that we can come up with the scale which is most convenient for us. We have heard such criticism from some sources, and that upset us. For example, in the ADNI study, 5 or 6 childs have been compared and verified. And what Eisai conducted, targeting MCI and early AD, we have verified against those clinical trials that Eisai has conducted in the past. So this is the scale, well verified through such initiatives. As a result, at early course of the disease, the specific items that are more likely to be impacted in earlier stages of the disease, those items have been verified as such and have been selected to compose this score: composite score [ CDR-SVP ], MMSE, ADAS-Cog, so -- since these items have been fixed up to constitute this score.

From [ CDRX Sum of Box ], all 6 items have been included. From MMSE, 2 items, and 4 items from ADAS-Cog. Those items that are more likely to be impacted in early disease have been included in ADCOMS, which is used in studies on other drugs by competitors, and it is at scale based on the verification. And we made those score -- items as combined together to make ADCOMS, which is our endpoint.

Again, next, I'd like to review the pathophysiology of the disease and profile of BAN2401. At CTAD, pathophysiology has been discussed. Amyloid is affecting tau pathway and leading to neurodegeneration. This flow of ATN to find a clue to the pathophysiology of AD, that has been the recent trend. BAN2401 is an antibody with a low affinity for amyloid beta monomers and higher affinity for aggregated amyloid beta species by over 1,000 folds.

Among aggregated species, large BAN2401 shows preferential activity for large soluble aggregate amyloid beta protofibril over fibrils by over 10 folds.

And BAN2401 neutralizes this amyloid beta protofibril and it creates aggregated amyloid beta species from the brain by [ Fc-region-mediated ] phagocytosis of the antibody. This is the mechanism of action. With this in mind, what will happen to the pathophysiology? And tau, which is another factor contributing to the development of the disease. In tau pathway, downstream of the pathway, nerve cells are damaged. This is something caused by amyloid beta, The phosphorylated-tau is one of the scale to assess that damage. And in CSF, phosphorylated-tau can be measured in CSF.

Furthermore, in neurodegeneration, to see whether neurodegeneration is taking place or not, then snaps damage -- synaptic damage and also degeneration of axons have to be looked at. And then for synaptic damage, neurogranin can be measured. For axonal degeneration, neurofilament light chain should be measured, and then the level of damage can be identified. This time, at CTAD, we presented data inclusive of all these factors.

Next as for APOE4 status. There are various discussions about APOE4 status. In Study 201, proportion of APOE4 carriers is 71% in placebo arm, and the number in 70s is consistent with the general ratio reported in amyloid beta positive early AD patients. And what we saw in this was, in placebo arm, APOE4 carriers and APOE4 noncarriers and overall arm were compared in the slopes, which represent the speed of worsening of clinical symptoms, result in placebo group APOE4 carrier and noncarrier subgroups. In terms of clinical decline, it was consistent with overall placebo group and were not statistically different from each other. Therefore, the conclusion is that APOE4 status is the risk factor for age of onset of disease, but has limited effect on disease progression at this stage of disease. That is what we can say from this data.

Next, second point about APOE4 status is about ADCOMS, what factors impact the worsening. [ Adapt ] data was also presented. The factors that will be examined is whether APOE4 status is positive or negative and whether it is MCI or mild AD and whether other AD treatment are given concomitantly or not and baseline ADCOMS. These 4 factors were considered. And this analysis follows mixed model repeated measures, which is a prespecified method. And this MMRM method is such that, for example, if there's a missing value, if there is a value that is not measured, then, with related data, a model can be constructed, missing value can estimated as a result and analyzed. That is the fantastic feature of MMRM. In pharmaceutical statistical analysis, MMRM is used widely, so it is not that MMRM is only used specifically for this analysis. And clinical symptoms worsening measured by ADCOMS was a test with MMRM to see what factors -- the earlier 4 factors were affected. And the p value of 0.05 -- below 0.05 means that there was an influence. But APOE4 status was not a contributing factor of P -- of 0.68. So APOE4 status is not a contributing factor in driving disease progression, but other factors were contributing factors.

The -- based on these 2 pieces of data, I think we were able to put a finish to APOE4 status discussion. At the highest dose of BAN2401, the slowing of progression measured by outcomes was not affected by APOE4 status.

Furthermore, in addition, then in the highest-dose group, when we looked closely at data on outcomes, what else can we observe? At the highest dose, the randomization base on APOE4 status is 48 APOE4 carriers and 113 APOE4 noncarriers. And APOE4 carrier ratio is, by coincidence, 30%. And the same comparison to 71% in placebo group, clearly, APOE4 carrier randomization ratio was lower. However, symptomatic -- less decline in symptom worsening. Also out of all group, APOE4 carrier was at 30%, but less decline was found to be 63% in APOE4 carriers and 7% in APOE4 noncarriers. And the first notion we had is that maybe severe APOE4 carriers are smaller in number. So maybe, the analysis was done on milder patients. But actually, it was the reverse. Regarding APOE4, the treatment effect is high. And APOE4 group was smaller, and this imbalance leads to underestimation of overall BAN2401 treatment effect.

Based upon regulatory authorities instruction, we were not able to continue to enter APOE4 carrier patients, but is -- it is consistent with about 70% APOE4 carrier, a percentage as in the general population. If we were able to randomize at that percentage, we may have been able to obtain even more robust data and may have been a possibility.

Furthermore, next highest dose, the 10-milligram monthly group, and a highest-dose group, where we pooled the 2 arms, and that data is shown here. By pooling these 2 highest-dose arms, APOE4 carrier percentage is 66%. It is not as high as 71%. However, it is more similar to general APOE4 status. And analyzing this pooled 2 arms, then overall is 21%, 25% for APOE4 carriers, 6% for APOE4 noncarriers, showing similar trends in terms of less decline.

And you might say that it is only sample size of 10 subjects. So a sample size you might think was very small. However, an 18-month time point partial for group analysis was not based only on 10 subjects. There were 48 randomized. And of them, there were patients on which administration of drug was completed. And such data was utilized using MMRM, and we made analysis based on 18-month data. And therefore, it is not correct to criticize that the sample size was too small.

In 10-milligram bi-monthly -- or biweekly and 10-milligram monthly, PK modeling suggests that Cmax is not the key driver but C average is a key driver of those response between 10 milligram per kilogram bi-weekly and [ 10 ] monthly. So this analysis supported the results that highest-dose arms showed higher treatment effect on 10 milligram per kilogram monthly arm. And this shows across the [ clever ] subgroup analysis results that changed from the baseline and the slowing down of declining of ADCOMS or improvement in clinical symptom decline on ADCOMS are shown. And there is consistently improvement in APOE4 noncarrier at 29% were clinically meaningful improvement in clinical symptom decline was shown. And therefore, in BAN2401, the mechanism to slow the symptom worsening is not only shown in APOE4 carriers, but also from APOE4 noncarriers.

To summarize APOE4 status, APOE4 status is not a contributing factor for treatment effect evaluation of BAN2401. The effect to slow cognitive decline observed in the highest dose arm is thought to be the treatment effect of BAN2401.

Next, at CTAD, BAN2401's possibility of a disease-modifying effect that was shown through data. And the first suggested disease-modifying effect is correlation of ADCOMS and PET SUVr data. Amyloid clearance in the brain, and this is a PET SUVr's change are shown, and this shows the slowing down of decline. And this correlation coefficient is 0.838. According to statistician, above 0.6 means that there is a strong correlation. So there is strong -- there is a correlation between ADCOMS and PET SUVr. In BAN2401, amyloid clearance in the brain and improvement in clinical symptom show the correlation. So this suggests the start and end of disease modifying where group correlation was shown.

And next is a slope analysis of ADCOMS. What did we look at from this analysis? BAN2401 highest dose, in that group, this shows the disease progression in highest dose from baseline. And in comparison to placebo, a significant slowdown of disease progression was shown. This p value is 0.001, so statistically, a significant difference was shown throughout the whole treatment period. This treatment effect was sustained. And it was also suggested that this treatment effect is potentially expanding over time, so this data is very important. Disease-modifying drug is different from symptomatic relief drug. It will remove the cause of the symptoms, so the defect is sustained, and effect grows over time. That is the condition to be a disease-modifier. So the possibility to satisfy that definition was shown through this data. That is how I look at this data.

Earlier, about pathophysiology, I discussed ATN. And please recall that ATN and T part. In T part, there is phosphorylated tau. This is a CSF marker. What this is, is that phosphorylated tau is related to tau pathology. And in downstream of tau, neurodegeneration is shown by this CSF marker. And in early Alzheimer's disease patients, high level of p-Tau is deducted in CSF. And in relation to tau, neurodegeneration is measured with p-Tau. And in BAN2401, in 18 months, there was a decline by 13% of CSF p-Tau level. So in a T part, there is a suggestion that this has positive effect on T part. And as for N part, neurogranin, neurogranin is synaptic protein. And it is a synaptic damage marker in CSF. And synaptic communication -- if other synaptic damage, neurogranin is increased. With the BAN2401, after 18 months, neurogranin level in CSF was lower by 11%.

And as for neurofilament light chain, it is a neuronal structural scaffold protein, and it is a CSF marker of axonal degeneration in neurons. Axon grows -- it becomes longer in neuron. And if there is axonal degeneration, neurofilament light chain will increase. And in 80 subjects, the CSF NFL level is elevated. But BAN2401 slows increase of neurofilament light chain in CSF by 48%. So this process, A and T and N, so the actions that disease modifier should have were exhibited.

To summarize, amyloid tau neurodegeneration regarding amyloids overall significant amyloid clearance confirmed after PET overall and over 80% of subjects have highest dose converted from positive to negative, confirmed through amyloid PET. With PET detection, these patients will no longer be diagnosed as Alzheimer's disease since they are amyloid negative. This is EPOCH making. And amongst the subgroups, amyloid clearance was observed in all of the subgroups.

In terms of tau, in tau pathway, a downstream -- there is a neurodegeneration, but the p-Tau decrease was observed, suggesting improvement in pathophysiology. And synaptic protection was also suggested because of production of neurogranin and a slowdown of increase in [ N cells ], suggesting inhibition of axonal degeneration. So in A, T and N, by lowering amyloid, we see positive chain reaction. And what is most important is clinical outcome and observe effect to slow cognitive decline in each valuation items in highest-dose arms, 30% less cognitive decline in ADCOMS was observed and the less cognitive decline was observed across all subgroups. And expansion of treatment effect is also observed throughout the treatment period.

To summarize, BAN2401 is such that disease modifying agent will help reduce the cognitive decline by potentially clearing out amyloid beta in the brain, which is believed to be a factor in the development of Alzheimer's disease. And observed effect or reduced amyloid beta aggregates in the brain, along with improvement of CSF biomarkers and slow clinical symptom decline, suggest potential disease modifying effect of BAN2401.

Regarding BAN2401, we have initiated interactions with major health authorities. In October this year, we have meeting with FDA in the United States. We had a productive collaborative meeting, as I was informed.

Regarding Study 201 data, to review that data in depth, additional interactions with FDA will be pursued since that was clearly outlined by the FDA, so preparations for such meetings, scheduling are underway. And we also have recently meeting with EMA, which was also held in productive, collaborative environment. And we soon expect to receive advice on future development in writing. As for meeting with the domestic Japan PMDA, it is planned in December 2018. And in order to collect the further clinical data, we have initiated open-label extension of Study 201. OLE of Study 201 was initiated, an additional confirmatory study to start early, we have started preparations.

Next, I would like to turn to elenbecestat, starting with profile. Elenbecestat has relative selectivity to BACE1. The ratio is 0.28. And substrate activity of BACE1/2 may be important for neuronal or synaptic maintenance and brain immunity. In this area, there are many development discontinuations. However, we are able to continue development of BACE inhibitor elenbecestat. And other based on development, but that is still continuing, also has a BACE1 selectivity. In preclinical studies, studies did not demonstrate serious adverse reactions, such as synaptic dysfunction, hypopigmentation and motor dysfunction. In Study 202, a dose of 50 milligram per day, that dose is used and this is a dose that will have more than 50% reduction of CSF A-beta, but not higher than 75%. So there is reasonable reduction of CSF A-beta, but not excessively high reduction.

Why are we using this dose? Why Icelandic mutation? You might have read a paper on Icelandic mutation, APP A673T variant, or so-called Icelandic mutation. Amongst those who carry this mutation, if A-beta production is reduced by 40%, A-beta -- Alzheimer's disease onset risk is reduced by 75%. So if we can achieve this, then we believe that, as BACE inhibitor, it will fulfill its objective. And that is how the dose effect. It is not excessively high. It is not excessively low. That is how this dose was selected.

Next, clinical assessment. Brain amyloid beta reduction was significant. Amyloid accumulation reduction was clearly shown. And as for clinical symptoms, CDR-SB and ADCOMS both were used. And in similar way, over 30% less decline was shown. Because of sample size, there is no p test. But amyloid beta reduction was observed under PET and increase of PET was significant.

And to summarize, risk-benefit balance was considered in this elenbecestat, which is a relative selective BACE1 inhibitor and, therefore, there is a differentiation from other BACE inhibitors. Because it was done with careful examination to balanced risks and benefits and the study results supported exploratory endpoints by showing significant difference of brain A-beta and trend of less decline on clinical symptoms. And initial impact on cognitive function seen in other BACE inhibitors, clinical studies were not observed -- was not observed.

At 50 milligrams per day, this was selected optimal dose setting by confirming balance of expected effect to inhibit A-beta production and safety. And this is the only dose that is used in Phase III study, MISSION AD study. In MISSION AD, there is data safety monitoring board. And this is done without our involvement at all. And interim analysis is done and safety is evaluated. In some cognitive functions on a case level, if there is a significant deviation, data safety monitoring board will be watching that. And already, 6 times a data safety monitoring board meeting was held. And there were no signals in terms of safety concerns. But about cognitive function effect, the board will continue to monitor. And MISSION AD is ongoing, aiming for full enrollment completion in fiscal 2018.

Next, aducanumab. A 36-month titration data was disclosed. In that -- about 36 months and 48 months, A-beta reduction in dose-dependent and time-dependent fashion was shown. And it was also suggested that there was less decline. And so we were able to confirm the data that was consistent from past findings. And there are ENGAGE and [ embrace ] 2 Phase III studies of aducanumab, and we have achieved full enrollment in July 2018.

As for lemborexant, this time, ISWRD POC was achieved. That is what I report -- that it what was reported at CTAD. This is not just simple insomnia. There is a circadian rhythm that is involved. And circadian rhythm is out of control, and that will be rectified. So this is a first-in-class drug.

On mild to moderate AD 60 patients, double-blind placebo-controlled parallel group study was conducted with actigram, which is favorable. Parameters related to circadian rhythm, nighttime sleep and daytime wake were evaluated over 4 weeks of treatment. And statistically, significant improvement in 24-hour circadian rhythm pattern was confirmed, so POC was achieved. This may become the first ever in the world ISWRD treatment drug.

As for insomnia disorder development, that is ahead of ISWRD development. The 304 Study, the second study, also resulted in good results, so we were successful in 2 clinical studies. Based on these studies, in the third quarter this fiscal year, we plan to submit in the United States and in the fourth quarter in Japan and would like to submit in other countries in fiscal 2019.

As for oncology, I would like to focus mostly on LENVIMA. First, LENVIMA in the past 6 months achieved a JPY 24.5 billion sales, increase of 66% year-on-year. China will be above the future. But in all regions, except China, we were able to achieve better results than the previous year. In all of the regions, we are cooperating with MSD and the United States collaboration is already underway. And in Japan, it is just starting. And in Asia and Latin America and Europe, collaboration is about -- will begin.

As for China, as you know, HCC indication received the designation of priority review in approximately 10 months after submission. Type B hepatitis is said to be a very important disease in China, almost a national disease. There is a very large number of patients who have type D -- type B hepatitis, accounting for about 50% of such patients in the world. And HCC due to hepatitis, which is quite intractable, it is said to be responding well to LENVIMA. So LENVIMA is much hope for drug in China as well. And we are preparing for launch of first in-house anticancer agent in the biggest market, 50% of world's HCC patients with Merck.

As for regulatory status, in Japan, we filed submission ahead of other country, and in the United States and in Europe and in other Asian countries. This is a simultaneous submission and approval achieved in Japan, U.S., Europe, China and Asia for the first time. Simultaneous submission and approval means that, in the same fiscal year, submission was done and approval was obtained. So from Japan, we are sending out LENVIMA for HCC therapy, so this is the drug from Japan. And we now have -- it's been 7 months after approval or launch in Japan and there is a huge significance of this.

In HCC treatment, Japan is #1. Diagnosis, treatment and prognosis management or outcome management, Japan leads other countries in HCC, so prognosis is vital. By far, Japan is the best for HCC. And we are able to accumulate experience and knowledge in Japan ahead of other countries. And after launch in 7 months, over 5,000 prescriptions were achieved.

And as noted here, there is a REFLECT study and ORR, please recall ORR, it was about 40% in REFLECT study. Tumor shrinkage was observed in about 40% of the patients. And in real world data, this was replicated, as presented in academic societies. Similar to the Phase III REFLECT study in real world, we are seeing similar results, so this is yielding a huge impact.

As for tumor marker, alpha-fetoprotein or thym [thymidine] kinase are tumor markers, and immediately after administration, decline is observed. And in image, tumor shrinkage can be observed by patient together with the patient's physician, so this helps motivate patient to continue treatment and gives joy that the condition is improving, and patient will be more motivated to continue with the treatment. This -- and for first-line, LENVIMA is given at early phase in terms of taste. And so after tumor shrinkage, it is possible to use a local regional treatment. So treatment algorithm itself may be changing as a result of LENVIMA.

And the biggest point about LENVIMA is AE management, adverse event management. The common adverse event is a decreased appetite, fatigue and malaise. And if there are such milder symptoms, a short withdrawal or dose modification may be conducted so that LENVIMA administration can continue.

And in case of hepatic encephalopathy, this is a transient phenomenon as a result of shunt-induced cirrhosis. So it is possible to continue LENVIMA administration with appropriate symptom management and dose modification. These are findings obtained in Japan, which we are sharing with other countries.

Now as for R&D collaboration in LENVIMA, in addition to current 3 indications, we are pursuing 5 additional indications, total of 8 indications, and all of these are combination therapy using LENVIMA and pembrolizumab. These are the lines of therapies, and what is showing in green is a designation of breakthrough therapy by FDA, first line, RCC and second line, endometrial cancer. These are designated as breakthrough therapies. In addition, basket trial is conducted to pursue exploratory possibilities. And by the first half of 2019, we have a prospect that we can start these studies. And especially for breakthrough therapy designation part, we would like to start as early as possible -- as soon as possible to obtain approval as soon as possible.

This shows our collaboration in commercial and medical aspect. And what we call first-base countries are shown here. These are major -- part of the major markets. And our medical, digital marketing -- sales force, medical, digital marketing, in all of these areas, we have started collaboration. Before the end of fiscal 2018 in all of these countries, we would like to start commercial and medical collaboration.

This is my last slide. This is a full year forecast. Revenue was revised upward by JPY 4.5 billion. And the cost is continuing to be on declining trend. And expenses will be expanded as scheduled. And therefore, total expense will be within the gross profit growth, so we have a strong financial discipline.

Operating profit was revised upward by JPY 4 billion to JPY 90 billion. And profit for the year attributable to all lines of the parent was revised upward by JPY 3.5 billion or JPY 3 billion. So operating profit was revised upwards by JPY 4 billion. And profit for the year attributable to owners of the parent was revised upward by JPY 3 billion. As a result, EPS is JPY 211.4. ROE is 10%. And dividends per share is expected to be JPY 150. ROEs of 10% is a financial target for fiscal 2018 [ gateway ]. And so we are achieving this 2 years ahead of schedule.

[Unknown Executive]

Thank you very much, [ Mr. Naito ]. We would now like to entertain questions. First, we would like to entertain questions from those who are in attendance in this hall, before accepting questions over the phone. If you have questions, please use the microphone, and please also give us your name and affiliation before you state your question.

Now if you have questions, please raise your hand. Yes?

[Unknown Analyst]

Only one question because of the time constraint. On 2401, your meeting with FDA, which was held in October, that was about the type B end of the Phase II. Is this the type of the meeting? And at the next meeting, what kind of meeting is it going to be, ADC with which title, an FDA? Or for only with Phase II? Do you think that they will accept the accelerated approval?

[Unknown Executive]

Neurology Business Group, Ivan Cheung is going to respond to your question.

[Ivan Cheung, executive]

Sure. The October meeting is the end of phase II meeting. At the conclusion of the end of the phase II meeting, the FDA outlined a series of meetings of additional information they would like to go in depth of BAN2401. As you can imagine, the end of Phase II meeting was only a 1-hour meeting. And it is a fairly collaborative effort with the FDA for the sponsor, Eisai, and the FDA to go in depth of the 201, aesthetic 201 data. And those are not specifically -- as you mentioned ADC, we're not specifically need to follow those procedures, so that's what we will be doing as well the rest of this fiscal year. And to your last question, of course, depending on how the additional meetings -- I wouldn't say meetings, I think a better way to describe those would be interactions. Those, in parallel with those additional interactions, of course, we also discuss about the next steps for BAN2401 going forward. At this moment, again, as you heard, we are -- the FDA and Eisai, we are in a fairly collaborative relationship right now, and we will be pursuing all the possibilities of going forward. I hope we can provide more updates going forward.

[Unknown Executive]

Thank you very much. Are there any questions? Person seated in the middle of the room, please?

[Kazuaki Hashiguchi, analyst]

I'm Hashiguchi from Daiwa Securities. About LENVIMA development timelines, including a time schedule, overall structure is shown on page 30. I think 1 box is 1 study. I think this is more or less fixed. So what will be the impact of profit and loss? LENVIMA R&D, you have received a milestone payment and you're using that reimbursement. So I think actual spending by Eisai is 0. And how long do you expect this to continue, that R&D for LENVIMA is 0 at Eisai -- the expense is 0 at Eisai?

[Unknown Executive]

On that question, Yasuno, who is responsible for partnership, will respond.

[Tatsuyuki Yasuno, executive]

I am responsible for clients with Merck. My name is Yasuno. Thank you very much for your questions.This is the development plan that is shown here and some of the studies will be carried ahead of schedule and other studies will take place according to the schedule. So in terms of impact on PL (sic) [ P&L ], when the contract was concluded, we had a plan, and we believe that it will be following that plan. And when advance payment will be made, I will have to clarify, but I think it will be after 2020.

[Unknown Executive]

Thank you very much. Next person in the front row, please pass the microphone.

[Fumiyoshi Sakai, analyst]

My name is Fumiyoshi Sakai, I'm from Crédit Suisse. On 2401, I have only one question. At CTAD, you have disclosed this information on the slide. APOE4, the carrier and the noncarriers were compared in the data in graph. And noncarriers, the symptoms were shown to be on the worsening trend, and I think how should we interpret this? APOE4 existence itself may affect the result of the study. So it may be confusing in terms of how to interpret the results.

[Unknown Executive]

Thank you very much for the input. From the Neurology business group, I would like to have Mr. Tsuno to respond.

[Masanori Tsuno, executive]

I am in charge of the clinical development. My name is Tsuno. I think you're pointing out ADCOMS CDI and ADAS-Cog are separated in the data, and only 1 in APOE4 noncarriers, there was a reverse trend. First of all, this is the Phase II study, but based on the adaptive design and more patients are randomized to the most effective dosage, but sample size is limited, so in subset, and that is this -- -- we cannot deny the fact that there will be some bumpy data or fluctuation in data. As regards to placebo arm, overall, APOE4 status is negative status, accounts for less than 1/3 of the population. So APOE4 noncarriers have operation to be analyzed, placebo arm is only 70 -- have 70, therefore, their sample size is very limited. And then APOE4 noncarriers, where active beta is increasing, however, the placebo arm data is decreased. Therefore, relatively speaking, there will be some imbalance. So please look at the data from the whole study. And what is most important is at the high dose and the high frequency, responders were observed. We analyzed the data, the clinical stage of AD and their concurrent drug use at a baseline value. And these were found to be very important factors all related to the results. And these were also factored in our analysis APOE4 status. Status itself, that did not show impact on the results. So considering that, and at the time [ of trying to ] study data, there was analysis on the subset of APOE4 status, and there was no difference as either. And there was no impact on the disease progression in placebo arm. That was explained by Biogen, inclusive of all these factors. APOE -- we would have been able to say that it is a very effective for APOE4 carriers, but irrespective of APOE4 status, our prediction is that this drug is efficacious. And as we have shown you the slide, the [ picks ] of the population analysis was also conducted, [ picks ] of the population. We were monitoring the tech result over the long time, as 18 months clinical data also accompanied these patients. Therefore, for those patients who withdraw, the data for such patients were not corrected. So that means that this data is trustworthy and then for such data, consistently 30% effect size was observed in the -- even in the APOE4 noncarriers. As has been shown this year, this today, combined data 10-milligram monthly data. In APOE4 carriers, there are so many. Total number of patients were 270, out of which about 220 were from 10-milligram monthly. Of course, half the dose means that the efficaciousness may have been reduced. However, even adding them 170 in placebo and active arm 270 patients, and then, we still had a 25% effect size. With half the dose, that means that the 80% of the patients were from the lower dose but still at such a high effect size was observed. At 10-milligrams monthly and bimonthly, I believe that the effect size would have been much higher. So in response to your original question, because subset analysis, which depend on the smaller sample size and [ intensive ] data captures have been improved, included therefore, there has been some bumpiness in the data sets.

[Unknown Executive]

Thank you very much. I'm afraid that we are not able to entertain further question. But maybe we can entertain 1 or 2 last questions?

[Shinichiro Muraoka, analyst]

I'm Muraoka from Morgan Stanley. I have a question about BAN2401. You're preparing [ retrospect ] additional validation study, what will be the study design? How long will it take? And when you discussed this, what was the reaction from FDA?

[Unknown Executive]

Thank you for your questions. That question, Mr. Ivan Cheung will address.

[Ivan Cheung, executive]

Cheung" type="E" />

Thank you for the question. The design is now under discussion with the health authorities. We will of course pursue a design quite similar to the 201 study because we have very, very high confidence of replicating the positive results in the Phase II study. Once we finalize the design, we will update everyone.

[Shinichiro Muraoka, analyst]

[ Two ] months at maximum?

[Ivan Cheung, executive]

Of course, this is a disease modifier. We will need to show longer term data. But again, the exact design, we will update everyone once we conclude the conversation with health authorities.

[Unknown Executive]

Thank you very much. I saw the 2 more hands, so 2 more questions.

[Unknown Analyst]

I'm [ Nibero ] from Tokio Marine Asset Management. LENVIMA, about HCC, I do not know when this is going to be, but if a checkpoint inhibitor is approved as a first line, can it be positioned as a second line in the case you have to obtain separate approval, or can it be used without additional approval?

[Unknown Executive]

Oncology Business [ Science ] responsible person, Dr. Owa, will respond.

[Takashi Owa, executive]

Thank you for your question. Let me repeat the question to make sure that I understand. Checkpoint inhibitor evolved from [indiscernible], let's say, nivolumab is approved as first line, then what will be the use for second line therapy drugs? I understand your question in this way. First, nivolumab, after it is used as a first line, what will we use as the second line? We have to collect real-world data and how drugs will be efficacious will have to be observed. And to do that, investigator-initiated study or cooperative-group study maybe necessary, not a company-led, sponsor-led study. And if data is collected and -- if robust data is collected, it will be included in the national guidelines. I think that will be the process. As for LENVIMA, before checkpoint inhibitor or after checkpoint inhibitor, it is effective according to the data so far.

[Unknown Executive]

Last question from the person.

[Motoya Kohtani, analyst]

Kohtani from Nomura Securities. Going back to Page 12. I think you didn't have to include this Slide, but 11 -- Page 11 can suffice. And on 12, I think the APOE4 carriers and noncarriers and I think that the error is overlapping. APOE4 carriers are positive or negative so you cannot tell whether there is any difference between the two only based on this data. [ What ] you can say is that the imaging and the test analysis are for sure, but what is the difference between BAN2401 and aducanumab? The only difference is the area, the systematic area. And the 34% for aducanumab and 10% for BAN2401. So where is this difference derived from?

[Unknown Executive]

Thank you very much. Mr. Tsuno, who is in charge of the clinical development, will explain.

[Masanori Tsuno, executive]

I'm in charge of the clinical development, my name is Tsuno. [ Area E ], as you pointed out, yes, there is a very big difference, particularly if you look at APOE carriers in aducanumab, I think it's about 56%, but in our case, it's only 14%. So why is it called [ the picks ]? Amyloid are better deposition, aggravation. The accumulation is the same. So that means that there must be difference somewhere else and this is between aducanumab and this drug is affinity. First, I do believe from a legal narrative that both of them have affinity, but our drug is relative high affinity to protofibril by over 10-fold and how the area is affected by this protofibril onset is not clear yet. I think this is a very big factor for this.

[Unknown Executive]

With this, we would like to conclude the financial results briefings for Eisai Inc. Company Limited.

Thank you very much. We would appreciate your continued support.