Accuray Inc (ARAY) 2017 Q2 法說會逐字稿

Good day ladies and gentlemen and welcome to Accuray Incorporated fiscal second quarter 2017 earnings call. At this time all participants are in a listen-only mode. Later we will conduct a question-and-answer session and instructions will follow the at that time. (Operator Instructions). As a reminder this conference is being recorded.

I would like to introduce your host for today's conference, Doug Sherk. Sir, you may begin.

Thank you, Terrace. Good afternoon everyone. Welcome to Accuray's conference call to review the financial results for the quarter ending December 31, 2016, which is the Company's second quarter of its fiscal 2017 year as well as recent corporate developments. Joining us today are Josh Levine Accuray's President and Chief Executive Officer, and Kevin Waters, Accuray's Senior Vice President and Chief Financial Officer.

Before we again I would like to remind you that during our call today the call includes forward-looking statements that involve risks and uncertainties including statements regarding our business plans and strategies as well as our outlook for fiscal 2017. There are a number of factors that could cause actual results to differ materially from our expectations, including but not limited to risks associated with the adoption of CyberKnife, TomoTherapy, Onrad, and Radixact Systems, commercial execution, future order growth, future revenue growth, and macroeconomic factors outside of the Company's control.

These and more fully described in the press release we issued after the market closed this afternoon as well as in our filings with the Securities and Exchange Commission. The forward-looking statements in this call or based on information available to us as of today's date and we assume no obligation to update any forward-looking statements.

During the question session we request that questioners limit themselves to two questions and then re queue with any follow-up questions. We thank everyone in advance for their cooperation with this process.

And with that out of the way I would like to turn the call over to Accuray's President and Chief Executive Officer, Josh Levine.

Thank you, Doug. Good afternoon, everyone and hank you for joining us on today's call.

Following the market close today we reported second quarter financial results highlighted by strong growth for both gross orders and backlog. Gross orders for the quarter increased 17% year-over-year to $78.5 million while our backlog grew 16% year over year to a record $426.2 million. We believe our progress is evidence that the catalyst driving order growth discussed with you over the last several quarters are beginning to take hold. These catalysts include a meaningful replacement cycle for Accuray products and greater recognition by both existing and new customers that our Company's product portfolio offers increased versatile and functionality for customers.

Over the next few minutes I'd like to provide more color on the most important of these catalysts before turning the call over to Kevin for a more detailed review of our financial performance and full-year outlook. First as many of you know entering the early phase of a multi-year replacement cycle of approximately 200 TomoTherapy and CyberKnife systems that will reach their ten year in service dates within the next 36 months. While we anticipate replacements will generally average 10% to 20% of orders on an annual basis the results in the second quarter were north of that range with a largest number of replacements occurring in the US market.

At the end of our first fiscal quarter we communicated we had secured commitments for several replacement orders in Q1 that were indeed finalized in the second quarter. We were optimistic about the replacement sales momentum continuing. When we look at the profile of our install base we have a mix that leans heavily towards large multi-bunker academic institutions, where many have our earlier generation Tomo and CyberKnife systems who would therefore benefit from the significantly improved functionality, versatility, and reliability of our newest devices.

Many of these installations are led by key opinion leaders who are engaged in research and recognize the value of improved workflow and efficiency associated with our latest generation products. Our view is bolstered when we look at installed customer base is that increasingly satisfied with the performance and reliability of our systems. Related to this, earlier in the month, we announced that Accuray Radiation Therapy Systems continue to set the industry standard for customer satisfaction.

Once again receiving MD Byline's highest composite overall user satisfaction rating among radiation treatment delivery systems in the US. In fact, most recent ratings report for California Q4 showed that Accuray radiation therapy products have now achieved the highest composite ratings among industry peers for 11 of the past 12 quarters. The MD Buyline survey results are consistent with the net promoter scores by IMV ServiceTrak where Accuray was rated number one in overall system performance.

We believe these improvements in system performance, reliability and ease-of-use are helping to drive gross order and backlog growth. Additionally Q2 order growth benefited from a improved order mix of our CyberKnife System. We continue to see the majority of the CyberKnife System orders configured with the multi-leaf collimator, which not only increases the versatility and speed of the system but also enhances the average selling price and margin profile.

While it's probably too early to attribute second quarter strength to recent clinical data we do believe the CyberKnife SBRT Study for low and intermediate prostate patients that was presented at ASTRO in September will enhance the strategic position and clinical validation of CyberKnife as an effective full body radiosurgery, supporting momentum in order growth going forward.

Additionally, while the majority of our Q2 CyberKnife orders in the US were replacement sales, throughout the rest of the world we saw a healthy mix of both replacement and new CyberKnife customers, which speaks to the expanded versatility of our latest generation CyberKnife platform. Transitioning to Radixact Q2 orders also benefited from the early rollout of our new Radixact System, the next-generation of our TomoTherapy platform, as we announced the first patient treatments had occurred in a number of our early reference site accounts in the US and Europe.

Customer feedback regarding the performance of the new Radixact System at those locations was extremely positive with regards to speed, efficiency and image quality of the device. Customers were also pleased with the performance of our new precision treatment planning software and it's adaptive planning and retreatment capabilities. Additionally in Q2, we continued our multi-system momentum with the booking of another multi-system order for three Radixact units with a prestigious Hong Kong sanatorium and hospital. Regarding Onrad, our new value product segment offering in China, we remain on track with regards to our initial launch.

As we have discussed previously regarding Onrad, a large element of the selling process is dependent on successful participation in provincial level tender processes. Based on the tender process timelines we expect initial order up take for Onrad to begin late in fiscal Q3 or early Q4 with further order ramp continuing from there forward. While we are pleased at the halfway mark of our fiscal year to be above our internal forecast for gross orders of backlog we recognize that potential variability in terms of outcomes and timing in two key areas potentially creates some revenue uncertainty in the second half of this fiscal year.

Let me touch on these two areas. Revenue conversion and the timing of Class A license issuances in China in more detail. First, we continue to see modestly expanded revenue conversion times due to a higher mix of international distributor orders in our backlog.

We've experienced an increasing trend in the percentage of our backlog related to distributor orders which increased 6% since fiscal year-end 2014. The impact of working with independent distributors in general is that we have less direct control over the timing and converting these orders to revenue versus our direct sales, process critical activity such as site planning, installations planning and coordination of contractors when the distributor has primary responsibility for system installation results in less direct line of sight with the end-user facility and limits our ability to manage all of the installation process, which can result on timing variability and delays.

The second area of potential impact is the timing uncertainty of a final announcement concerning Class A radiotherapy licenses from the Chinese Ministry of Health. Given the delays in this announcement over the past several quarters it's becoming more difficult to predict when the Class A license program specifics will be finalized. Our internal assumption supporting our second half fiscal 2017 revenue forecast is tied to a final announcement regarding Class A radiotherapy licenses coming by the end of our fiscal third quarter.

But given the delays we have seen and the the uncertainty of timing going forward we believe it's prudent to remind you of the potential impact that our second half revenue outlook related to continued delays. With that said we remain very confident in the China market opportunity overall and our competitive positioning in the Class A radiotherapy segment.

And with that I would like to turn the call over to Kevin.

Thank you, Josh, and good afternoon, everyone.

We ended the second quarter with back log of $426.2 million, up 16% over the prior year. Backlog included gross orders of $78.5 million and net orders of $54.1 million, reflecting age outs of $19.9 million and one cancellation for $3.6 million. Age outs in the quarter were anticipated and consistent with the guidance we provided in October. In addition currency negatively impacted net orders by approximately $1 million in the second quarter.

Josh already highlighted the factors contributing to our strong growth order performance including replacement sales above historical averages, the improved mix of both new and replacement CyberKnife, and the multi-system Radixact quarter. All of these items led to significant gross order growth in the quarter and we are approximately $10 million above our guidance for the first half of fiscal 2017. First half of fiscal 2017 gross order performance translated into 16% year-over-year growth in our backlog.

As we look out 12 to 24 months, when these systems are installed, our backlog could translate into product revenue growth rate greater than our rate of product growth in fiscal 2017. Turning now to our income statement. Total revenues for the second quarter were $87.5 million compared with record levels of revenue of $108.9 million in the year-ago period. On a sequential basis revenues were up slightly over the first quarter and we continue to anticipate a linear progression of revenue as we move through the balance of the year.

I'll provide more detail on this metric shortly. Product revenues for the quarter were $35.4 million, we performed above our expectations for the quarter in the US and Japan. This was offset by below planned performance in Europe which is primarily attributable to the modestly extended conversion to revenue times that Josh highlighted.

Service revenues for the second quarter were $52.1 million compared with $53.2 million in the year-ago period, which was in line with our expectations. Turning to gross margins, our overall gross margin for the quarter was 36%, down 320 basis points compared with the prior-year period and flat on a sequential basis. Product gross margins in the quarter decreased 620 basis points over the prior-year period to 35% primarily driven by the lower overall unit sales volume as well as product and channel mix.

Product and channel mix continue to be the most significant factor in regards to quarterly fluctuation in product margins. As with the first fiscal quarter we again experienced a higher creation of TomoTherapy systems to revenue as compared to CyberKnife. In prior year we averaged an approximate one-to-one ratio of CyberKnife to TomoTherapy revenues and in the first six months of 2017 this mix has been two to one in favor of TomoTherapy.

I would also highlight that overall pricing in the first half of 2017 has been relatively stable to the year-ago period and within our expectations. We continue to anticipate an improvement in product gross margin in the second half of the fiscal year on higher revenue volumes and more favorable mix that should allow gross margins on a full year basis to be approximately flat as compared to the year-ago period.

Service gross margins in the fiscal second quarter of 36% were essentially flat with the year-ago period. We continue our focus on improved parts reliability and cost management and anticipate full year fiscal 2017 service margins to be at or above fiscal 2016, which were 36%. Operating expenses for the quarter decreased 15% from the prior-year period to $36.2 million.

The $6.5 million decline included a $3 million decrease in R&D, a $1.2 million decrease in selling and marketing and a $2.3 million decrease in G&A, primarily due to lower legal expenses. A contributing factor behind the lower operating expenses across all categories was incentive compensation and commission expenses as our result of a lower sales volume. Adjusted EBITDA was $1.8 million compared to $6.8 million in the year-ago period.

We continue to have a particularly high level of confidence in our ability to effectively manage costs, which I'll touch on in our guidance shortly. Let's now turn to the balance sheet. We had $108.4 million of cash and investments at December 31, 2016.

The $16 million sequential decrease in cash from the first quarter is primarily due to the timing of revenue collection, which can be seen in the $14.7 million increase in accounts receivable. Additionally, we used $6 million to pay down current debt balances. Turning to our capital structure. We remain focused on strengthening our balance sheet by reducing the potential dilution associated with our convertible debt.

We will report to you on our progress as developments merit and we continue to believe Accuray's annual financial performance and growing backlog will provide greater options and flexibility. Turning now to our annual guidance for fiscal 2017. We are, today, affirming the guidance originally provided back in August for all metrics except operating expenses as follows. Year-over-year growth of approximately 5% for gross orders, a revenue range of $410 million to $420 million.

Although our first half performance leads us to believe the lower end of the range is more likely and dependent primarily on two variables that I'll explain shortly. We are expecting adjusted EBITDA range of $32 million to $38 million. Additionally, given first half operating expense levels were below expectations, we are lowering full year operating expenses to be down approximately 3% as compared to fiscal 2016.

Given the lower operating expenses we do remain confident in our ability to achieve our adjusted EBITDA range, which would represent year-over-year growth of between 30% and 55%. With regard to full year revenue implicit in our revenue guidance are two variables that will impact our ability to achieve the revenue range. Josh mentioned our internal assumptions expectations of a Class A license announcement out of China in the third quarter.

China Class A revenues comprise approximately $10 million to $15 million to our second half fiscal 2017 revenue guidance. Additionally, almost all of the China revenue contribution is now anticipated to be in our fourth quarter. Our revenue guidance also anticipates the shipment or installation of several distributor orders within the time frame inspected when these orders were entered in the backlog.

At this time we have no reason to believe the conversion of these orders to revenue will experience timing delays. As with China our firm belief is that the conversion to revenue is purely a timing issue. Further, we would anticipate closing out fiscal 2017 with product system backlog of approximately $440 million, which should be above the overall market growth rate.

We also anticipate Q3 age outs to be approximately $12 million. We anticipate that age out, as a percentage of backlog for the full year fiscal 2017, will be an improvement over fiscal 2016's level of 13%.

And with that I would now like to hands the call back to Josh.

Thanks, Kevin. Before we open the call-up to your questions I would like to thank the entire Accuray team for their focus and contribution to our performance in the second quarter. Our Company's commitment to the important work we are doing in improving the lives of cancer patients by providing leading edge treatment solutions remains central to our mission.

Now, Operator, we're ready to open the call up for questions.

Thank you. (Operator Instructions). And our first question comes from Josh Jennings from Cowen and Company. Your line is open.

Hi. Good evening. Thanks, gentlemen.

I was hoping to ask a few questions on the sales funnel. Clearly the replacement cycle has set up nicely. Have any high-level thoughts on how you're feeling about the de novo customer funnel at mid-year versus the beginning of the year? Then on Radixact, you know, one of the promise of the platform that it will more fully open up the single and dual vault segments of the market. Can you talk about the early Radixact funnel and the mix of single and dual vault centers within it? And then lastly, you secured another multi-system order in fiscal Q2. Does the funnel include more potential multi-center -- or sorry, excuse me, multi-system orders as well?

So, Josh, I'll take -- I guess in that -- that sequence of questions were given. Let me talk a little bit about replacement cycle in general. As we having characterized in the past the primary geographic focus is US market and Western Europe across both platforms.

If you think about just the performance let's say in Q2 and the contribution -- the bigger contribution in terms of product mix from a CyberKnife standpoint, you know, right now across the entire install base, that's coming into replacement cycle on CyberKnife, less than 20% of that install base actually has the latest generation device. So when you think about the opportunities for M6, MLC equip, people that are looking to build full body radiosurgery programs we think we have a lot more ground to cover in terms of opportunity inside the install base and that -- that's represented by replacement cycle.

Let me talk a little bit about product advantages on both platforms and we have -- the basis here for confidence really comes from a better performing product on both platforms. Better functionality, better speed, better throughput, better reliability and those things really translate into expanded clinical versatility in the minds of customers but also a growing addressable market opportunity for us.

So I'm pretty bullish on the continuation of momentum in the replacement sells piece in both the US market and Western Europe.

I can take your single and dual vault question. We continue to experience the majority of our orders for Radixact and TomoTherapy in single dual-vaults which has region been running now north probably the last four or five quarters, Josh, so that trend continued in the second quarter.

Josh, I think the last question you had was around Radixact and kind of funnel and what our view is right now from an early assessment. We are in -- still in this current quarter ramp and monitor, which is really kind of a reference site installation optimization exercise that essentially make sure we're creating the very best reference sites we can and understanding elements of installation that are specific to the new platform. That program is not completed until the end of this quarter so the fiscal third quarter but, quite frankly as you can see we're not -- it's not standing in the way of us take I guess orders for Radixact.

So we have got sites in both the US and Europe that are reference sites, that we have already taken to revenue and that are all treating patients today. And, again, we expect that they'll be completed with ramp and monitor as a phase by the end of the third quarter but generally the feedback from those customers that now have patient treatment experiences with the new system has been really positive. Very strong feedback -- positive feedback about speed, product performance.

In some instances exceeding their expectations quite frankly so we have -- we have, again, a stronger sense that we have got with Radixact a true workhorse product and -- and a product that, quite frankly, is going to really meet the needs of a lot of customers.

Thanks for that. Sorry for the multi-prong question. Just one follow-up. Just on the replacement cycle. I'm sorry if you guys commented on this in your prepared remarks, but can you give us an idea of whether you're maintaining that replacement conversion rate? And then lastly with Radixact and with the next-generation of CyberKnife, with the (inaudible) MLC, are you seeing any customers move up their replacement timeline to have the latest and greatest Accuray technology platform earlier than say ten years? Thanks for taking the questions, guys.

So I'll start. I'll let Josh come follow up on the timing. We did see, in the second quarter, Josh, above I would say historical performance on the replacement cycle.

We have said that you should expect in fiscal 2017 for us to be in the 10% to 20% range as a percent of total orders. We were north of that in the second quarter, but I don't get too excited about looking at it on a quarterly basis because I think for fiscal 2017 we're still comfortable with replacements being somewhere in the 10% to 20% range of total orders.

Josh, on the accelerated timing, if you will, part of the question, you know, if you -- again, if you look at the profile of the typical location that we have our install base equipment in, you're talking about large multi bunker academic institutions, generally a more complex case mix of patients that end up being treated in those institutions and, in many cases, people that lead those departments and those -- you know, the radiation oncology practice, if you will, at those medical centers are pretty well known key opinion leaders, people that are still actively involved in research with those patient populations and our devices.

So there's generally a desire to have, what I would characterize as, the latest and greatest in terms of our equipment. I think maybe in certain cases you see people that would -- would, you know, and have the capability from a budget and a flexibility standpoint to be able to go to upgrade perhaps sooner than the ten year reinstatement cycle, but our expectations are that it isn't necessarily that that's going to happen on a routine basis or a highly predictable basis.

We've got -- we've got a lot to keep us busy, quite frankly, just at the level of replacement sale opportunity in front of us. Again, it's 200 plus systems across CyberKnife and TomoTherapy on both platforms and, you know, again, I think we have a fairly robust view of the momentum, if you will, of replacement sales continuing.

Thanks for all those answers. Appreciate it.

Thanks, Josh.

And our next question comes from Anthony Petrone from Jefferies. Your line is open.

Thanks and good afternoon. Maybe I'll start with a higher level question just on the market landscape and then move into a couple on guidance, but maybe just, Josh and Kevin, if you can provide an update on the competitive landscape, you know, just as it relates to Elekta and Varian and maybe the similarities and differences and the landscape heading into calendar year 2017.

So I'll start, Anthony. I think, again, if you -- if you look at where certainly where our replacement sales opportunity resides, again, you have got heavy creation in concentration in the US and Wester Europe. In the US, you know, the primary people we're competing against is Varian. Probably seven out of the ten bunkers that our sales people drive past have a Varian device of one form or another running in those vaults.

So you know it is a very -- I'd say it's a very from a competitive scenario it's a very Varian centered kind of landscape. I think we're doing a really good job of maintaining and retaining our existing bunkers and we are showing that with improved functionality and product capability and reliability we have got the -- you know, in a competitive sense we have got the ability to convince, you know, single and dual-vault locations that Tomo and now Radixact are really viable options for them where over time if you look back over the last four or five years, that -- that quite frankly historically wasn't the case.

So I think we're -- I think we're feeling pretty good about the competitive dynamics, again, we're not -- we're not winning at a hundred percent. We're not going to win at a hundred percent, but I think we're being a more viable competitor, quite frankly, today than we have been at any point in time in my experience at Accuray.

That's helpful. Maybe just to clarify from a bunker size standpoint is Tomo and Radixact similar to, sort of, the existing dimensions for the majority of those installed Varian bunkers?

Yes. They would -- they would certainly be more characteristic of a -- you know, a relatively easy substitution -- relevantly minimal bunker modification would be required so that certainly helps d helps the discussion.

Okay. And then just on guidance you mentioned distributor orders. I'm just trying to get a sense of the difference in timing of revenue recognition, you know, sort of when of a distributor order versus a direct sales order and maybe just the last question would be, you know, what do age outs and cancellations look like with respect to distributor orders versus directs orders? Thanks.

Yes. So I'll answer your age out question first. You know, our age out historically have been primarily distributor orders as compared to direct orders.

With that said, the comments I made about expecting age out in 2017 to improve as a percentage of backlog over fiscal 2016's level of 13% takes into account the increase in the revenue converse cycle that Josh and I highlighted. So we've already factored in any impact that would have in our age out guidance. Your first question in regards to differences in converse timing, I mean, the fact that it does vary by region and you're going to see some differs.

I think in this quarter we are impacted primarily, as I mentioned, on the revenue side by our distributors in the European region primarily the none Western European regions where we use distributors.

However, on a positive note we highlighted I think in our third quarter call last year Japan which is a where we are a hundred percent distributor we have actually seen that converse process improve over the last six to nine months. So much so that Japan now in the first half of 2015 is representing approximately 15% of total Company revenues as compared to 5% in the first half of 2016. So it's very region specific, macro factor specific, but the weakness in the delays we saw in the first half of 2017 were primarily Europe.

Very helpful. Thanks.

And our next question comes from Tycho Peterson from JPMorgan. Your line is now open.

Hi, guys. Thanks. This is actually Patrick in for Tycho. Maybe just to expand on Anthony's question on revenue the converse. You know, I definitely understand it's more difficult to handicap give distributors relative to direct control but now that you have kind of identified it as a problem, have increased focus on it, is there anything you can do to accelerate conversion from the distributors or just how are you thinking about improving that a little bit?

Yes. So, Patrick, it's a great question. Just a bill bit of background on the resources that we usually have involved when we're involved in a direct sale and then where we have primary responsibility for installation.

We've got installation teams and essentially revenue converse teams in our regions that are working with end-user facilities on-site planning elements and site planning details, working on installation details and bunker modification components. You know, elements of the process, and coordination, quite frankly, in some cases with contractors locally to support those activities that I just described. Those resources are -- are our resources are dedicate and they're affect focused by region and when we're a directs sales mechanism and we've got responsibilities for installation those resources are highly engaged, actively engaged in a really end to end process with the site and the institution end-user institution.

We are trying it to work more closely with distributors to find out -- it's really a bit of the help us help you discussions. We're working with them to find out what we can bring, with regards to some of the resources I just described, that would help augment or help support perhaps in ways that they're not capable of on their own those -- you know, the execution, if you will, of those activities that I just described. So, you know, we have got a I large number of independent distributors that we work with in different parts of the world, but this is not necessarily an exercise for us where we've got to pore every one of them.

The orders are probably more heavily concentrated in, again as Kevin described before, Central Europe. Certainly as of most recently Japan, all that's improving situation, as Kevin highlighted. So we have -- I think we have the resources from the Accuray side that need to be engaged in these conversations with the distributors pushing and -- and expanding their involvement perhaps, if you will, in efforts to move -- move this process forward in a more expedited fashion.

Okay. That's helpful. And then maybe one for Kevin on gross margin, you know, they we obviously a bit soft in their first half. Looking out you guys gave at the Analyst Day longer-term target 45%. Can you just talk to your confidence level in that number and the key levers to get obviously a lower base coming off the first half here.

Yes. So as I mentioned your question really was two part. On the second quarter, as I said, the most significant factor continues to be really mix and we were hurt in the first half of the year by product mix, having a higher ratio of Tomo revenues, and also just an overall lower sales volume.

We still think given our forecast for the back half of the year we can get back to essentially flat product margin in 2017 as compared to 2016. But longer-term we still continue to work towards the goal we put forth on Analyst Day of 45% overall margins. This would represent service margins in the low 30%, 40% range, product margins in the high -- in the high 40's.

Revenue improvements are an obvious factor in regard to margin improvements. But another thing we're also working on as a company, which we haven't talked about in a few quarter, we do remain focused on cost down initiatives on our systems and we are currently investing in several large -- large sustaining engineering products which whether not benefit fiscal 2017, but it is going to contribute as we head into 2018 and 2019. So I don't view our overall margin results in the first six months as any risk or indication that we can't get to where we want to get to in fiscal 2019.

Understood. Maybe just one last one, you know, you guys mentioned Japan as an increase in part of revenue. I don't think you have shown a (inaudible) for Radixact yet. So maybe, when are you anticipating that, what's the visibility there on that? And then how significant the opportunity for Radixact? Thank you.

So we have submitted all the necessary paperwork for (inaudible) approval in Japan and we expect that very shortly. We expect it in our fiscal third quarter.

(Operator Instructions). And our next question comes from Brandon Henry from RBC Capital Markets. Your line is open.

Yes. Thanks for taking my question. Motion management, it's an important differentiator for CyberKnife that is not yet available on the Radixact platform, so can you provide us with update on the timeline for motion management for Radixact and then discuss how important you think this technology will be in driving TomoTherapy replacements and I have a follow-up question.

Brandon, it's a great question. What you described in terms of capability on CyberKnife, which is the motion management capability, the software component to that which is the automated beam adjustment capability through synchrony. These are similar technology upgrades that we believe are going to be really pretty -- pretty central to longer-term adoption of -- and utility of Radixact going forward.

You know, this is a highly upgradeable platform that will allow for motion -- motion tracking over time, motion correction in terms of beam correction over time. So it's -- if you think about how CyberKnife has evolved, Radixact would look -- again, different platform obviously but would look similar in terms of how it tracks along those upgradability pathways, if you will, and this is a major at least by feedback this is a major part of the thought process for those customers that are, you know, looking at considering Radixact in -- in the replacement sales discussion.

Again, if you think about it in terms of our -- our existing install base, we're taking a -- we're taking a -- you know, a customer with earlier generation product and even in an upgraded form that earlier generation product won't ever get to the level of capability from a functionality perspective that Radixact will given the functionality improvements that I have just described.

And we haven't been specific quite frankly about timing but I would say that we feel pretty good about the rate at which -- or the -- the pace at which we're moving through our technology roadmap and advancing the development of these -- these incremental product capabilities.

Okay. And just a couple other questions. First, have you seen any differences in the historical win rate for replacements for TomoTherapy versus CyberKnife and then also the replacement rates for the US versus international market?

No. We haven't. I think the answer would be it's too early to draw any definitive conclusions.

You know, we're very early in this replacement cycle. I will say the first half of 2017 yielded positive -- positive results on both product platforms. We saw in the US predominantly all the replacement sales and those were predominantly CyberKnife related while in the OUS we saw a nice mix of replacement sales and competitive wins across both platforms.

So I don't think -- I think it's too early in the cycle, Brandon, to draw any conclusions. We're performing at or above expectations on replacement sales across both product platforms at the moment.

Okay. Then final question for me. You mentioned the operating expenses will now be down 3% to 4% for this fiscal year versus I think previous guidance was kind of flat.

Can you talk about where the cost reduction is coming from and also if -- if fiscal second half revenue performance does fall short of expectations are there additional opportunities for cost cutting that would sill allow you to meet your fiscal year 2017 adjusted EBITDA guidance? Thanks.

Yes. I believe so. So we finished the -- for the first six months of 2017 roughly down $10 million in OpEx or $12% below prior year, $4.4 million roughly $4.5 million of that is attributable to legal expenses, I'll call them non-recurring.

And what you're seeing in the first half is lower revenue levels leading to go decreased compensation and decreased commission expenses and our guidance right now of operating expenses down 3% assumes we're within our revenue range of $410 million to $420 million and if we would be below that range, I think it's fair to suggest that our operating expenses could potentially be down less than 3%, primarily due to the two factors on compensation that I just mentioned, which, by the way, in my prepared remarks I had said that's what gives us I would say a particularly high level of confidence in our ability to achieve our EBITDA range.

Okay. Thank you.

No problem.

And at this time I'm showing no further questions. I would like to turn the call back to management for any closing remarks.

Thank you, Operator, and thank you everyone for participating on this afternoon's call. We look forward to talking with you again on our fiscal third quarter earnings call and thanks and have a good evening.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program. You may now disconnect. Everyone have a great day.