Accuray Inc (ARAY) 2019 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q1 2019 Accuray Incorporated Earnings Conference Call. (Operator Instructions) As a reminder, this call is being recorded.

I would now like to introduce your host for today's conference, Todd Kehrli with EVC Group. Please go ahead.

Thank you, operator. Good afternoon, and welcome to Accuray's fiscal first quarter earnings conference call. With me on today's call are Josh Levine, Accuray's President and Chief Executive Officer; and Shig Hamamatsu, Accuray's Interim Chief Financial Officer.

Before we begin, I'd like to remind everyone that, on today's call, management will make forward-looking statements that involve risks and uncertainties, including statements regarding our future results as well as business plans and strategies. 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 our products, our ability to develop new products or enhance existing products, commercial execution, future order growth, future revenue growth, future margin expansion and macroeconomic factors outside the company's control. These and other risk factors are 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 on this call are based on information available to us as of today's date, and we assume no obligation to update any forward-looking statements.

Two housekeeping items: first, (Operator Instructions) Second, all references we make to a specific quarter in the prepared remarks are to our fiscal quarters. For example, statements regarding our fiscal first quarter refer to our fiscal first quarter ended September 30, 2018.

With that said, I'll now turn the call over to Accuray's President and Chief Executive Officer, Josh Levine.

Thank you, Todd. Good afternoon, everyone, and thank you for joining us on today's call. Before we begin, I'd like to introduce Shig Hamamatsu, our Interim CFO, who is joining me on the call today. Shig was our VP of Finance and Chief Accounting Officer before assuming the role of Interim CFO on October 1. Shig has done a great job as the Interim CFO during this transition.

With regards to Q1 performance, we had a solid start to our fiscal year 2019. First quarter gross orders of $61.4 million were up 10% year-over-year, with especially strong demand for our Radixact System. Gross orders for our Radixact System more than doubled year-over-year, receiving approximately 25 system orders highlighting the growing market interest and adoption of our newest generation treatment offering. The strength in Radixact orders also highlights our expanding customer base with more orders coming from community and regional-based hospitals.

Breaking down gross orders by type, approximately half of our Q1 orders for new vaults. We also had a strong quarter for competitive replacements, which more than doubled year-over-year and represented approximately 40% of our gross orders for the quarter. Again, this activity was driven by our Radixact product line. It's important to note that these orders from both new and competitive vaults represent positive indications for future installed base growth.

Lastly, 10% of our gross orders were replacement sales for existing Accuray Systems. We believe that the success we are seeing with Radixact in new vaults as well as with competitive replacements is because of improved performance characteristics related to treatment speed and overall throughput. These characteristics have positioned Radixact as a highly efficient and versatile platform, exceedingly capable of treating routine cases with great speed and efficiency, while allowing for complex case treatment capability. Radixact's functional performance and efficiency improvements, early positive customer feedback and our upgrade road map have all contributed to a much stronger product position for Radixact across all customer profiles.

Gross orders for our CyberKnife platform were down year-over-year primarily due to quarterly variability in order timing. As a reminder, we reported historically high CyberKnife orders in the fourth quarter of fiscal year 2018. Regardless of the quarter-to-quarter variability, given the positively received launch of our CyberKnife VoLO optimizer software last week at the ASTRO meeting, which substantially improve system productivity with both increased patient throughput and reduced treatment planning time, we still expect to see full order growth on CyberKnife platform.

Turning to our regional performance. We saw continued growth from our EIMEA region, which accounted for almost half of our gross order dollars for the quarter, with a significant contribution from our distributor-managed markets. While we continue to gain traction in our distributor markets, we also continue to see successes in India where we are selling direct. Overall, the EIMEA region grew 10% last fiscal year and continues to perform very well for us generating double-digit order growth, again, in the first quarter of fiscal 2019 relative to the prior fiscal year.

In our Asia Pacific region, while we continued to generate orders in China in Q1, the big China-related news is the long-awaited announcement regarding Type A and Type B radiotherapy quotas and licenses, which were released by the Chinese Ministry of Health last evening at roughly 7:30 p.m. Pacific standard time. The announcement confirmed that there will be licenses issued between now and the end of calendar 2020 for 188 Type A radiotherapy systems and 1,208 licenses issued for Type B radiotherapy systems. While we are very pleased that we have final confirmation on this topic based on both the timing of this late breaking news and the depth of analysis required, we will not be giving any details on this call regarding estimates or guidance adjustments for FY '19 related to last evening's announcement from China. We expect that when our analysis is complete, we will be communicating in great detail about the potential benefits to Accuray from the release of Type A and B radiotherapy licenses.

Additionally, nothing related to last night's announcement regarding quotas and licenses has changed our strategy regarding establishing a joint venture in China with a local partner, and we are continuing to advance our efforts related to this objective.

As we continue to work towards achieving this goal, we will provide updates as developments merit. Regarding tariffs in China, we saw a modest pricing impact based on orders received in the first quarter, but generally we don't expect significant margin erosion from a system standpoint given historically high end user pricing levels.

In our Japan region, we continue to see solid order generation for our Radixact Systems in Q1, with a strong contribution this quarter from competitive replacements representing half of the orders we received in Japan. We were down a few orders in Japan year-over-year primarily due to quarterly variability and timing, and we expect our orders will increase from this region for the remainder of the year. Since receiving Shonin approval last year for our integrated data management system and our Precision Treatment Planning software, our funnel in this region has continued to expand.

In our Americas region, our initiatives to improve our growth in the U.S. market yielded positive results in fiscal 2018. We continue to focus on key initiatives, including increased attention on replacement of Accuray systems -- existing Accuray systems, multisystem opportunities and competitive wins. While Shig will provide a detailed review of the P&L in his prepared remarks, I wanted to provide a few financial highlights for the quarter.

Total revenue for the fiscal first quarter increased 5% year-over-year to $96 million. Product revenue for the first fiscal quarter increased 7% year-over-year driven primarily by the increased adoption of our Radixact System. In the first quarter, we also generated adjusted EBITDA of $4 million. As noted in today's earnings press release, we announced that we have initiated a strategic organizational restructuring that will reduce the overall cost structure of the company, while preserving our ability to continue our product innovation objectives and generate higher levels of sales growth going forward.

The workforce reduction actions taken, which were focused primarily across noncommercial functions will make Accuray a leaner and more market-responsive organization. We estimate that these actions will result in annual savings of approximately $15 million. We expect the full benefit of these cost reductions will be realized in the current year fiscal fourth quarter, and we believe these actions will give us a clear path to GAAP net income, while we continue to focus on top line growth. We believe that achieving a positive GAAP net income position will allow Accuray to generate improved levels of cash, continue to pay down debt and unlock business development and investment opportunities that were previously unavailable to us.

Before I turn the call over to Shig, I want to touch briefly on the innovation and product advancements we highlighted at the annual ASTRO meeting last week in San Antonio. As I mentioned last quarter, one of the most important development programs for our Radixact System is the launch of our motion tracking correction and beam synchronization capability, which is similar to the Synchrony system on the CyberKnife platform. We're pleased to announce that during the quarter, we submitted our 510(k) application for similar functionality on our Radixact System.

Last week at ASTRO, we previewed how this proprietary technology will work in our Radixact System and the reception from existing and prospective customers was outstanding. Similar to our CyberKnife platform, our proprietary Synchrony target tracking and motion correction capability for Radixact corrects for target movement during respiration and ensures the beam is synchronized with movement of the target.

Based on the projected turnaround timing on our 510(k) submission, we estimate that the first customer installation of Radixact with motion tracking correction and beam synchronization capability will be treating patients by the end of our fiscal year.

As mentioned earlier in my remarks, at last week's ASTRO meeting, we also previewed the CyberKnife VoLO optimizer. Those who had a chance to see -- use this software have been very impressed. The CyberKnife VoLO optimizer reduces treatment delivery time by as much as 50% and reduces treatment planning development time by as much as 90%. Plans created using the CyberKnife VoLO optimizer are also reported to improve plan quality even further than observed in earlier generations. In conjunction with our Precision Treatment Planning software, we believe that CyberKnife VoLO optimizer is the latest in a long line of next-generation functionality and performance improvements that position the CyberKnife as a fast, efficient and remarkably precise full-body treatment platform for stereotactic radiosurgery and stereotactic body radiotherapy.

We expect these improvements will be catalysts in driving continued installed base upgrades and full year order growth for our CyberKnife platform.

Lastly, on the topic of products and innovation, we recently announced the published data from 2 of the largest multicenter studies we've referenced in the past focused on SBRT prostrate treatment utilizing CyberKnife. The published data reinforced that the CyberKnife system provides excellent prostate cancer survival rates in 5 or fewer treatment sessions. In fact, for low-risk patients, the disease-free survival rates were between 97% and 100%, which was superior to the 92% to 94% from historical conventional radiotherapy data. These studies also reported superior survival rates for intermediate-risk patients. We believe that the results of these studies further validate the superior clinical utility and efficacy of our CyberKnife systems in treating prostate cancer patients.

I'll now turn the call over to Shig, who'll provide more details regarding our fiscal first quarter results as well as update our fiscal 2019 annual guidance in light of the cost reductions I just discussed. With that said, I'll turn the call over to Shig.

Thank you, Josh, and good afternoon, everyone. As Josh highlighted, we had $61.4 million of gross orders in the quarter, representing an increase of 10% over prior year. Included in our gross order are $1.1 million of upgrade orders purchased through our service contracts, which as we previously communicated are included in our gross orders beginning in fiscal year 2019. Excluding these upgrade orders, we had $60.3 million of gross orders, representing an increase of 8% over prior year. Both APAC and EIMEA regions were the drivers of the strong gross order growth. On a product mix basis, the first quarter was highlighted by continued momentum for the Radixact System, which more than doubled year-over-year. Additionally, Radixact represented approximately 85% of TomoTherapy platform orders.

On a net basis, we generated $24.9 million of orders, which included adjustments of $26.5 million for age-outs, $6.6 million of cancellations and $3.4 million for other adjustments. In the first quarter, we had 4 cancellations, which by geography, were in Japan, Latin America and Europe. Net age-outs for the quarter were primarily driven by orders related to China Type A systems, which represented more than 40% of the age-outs. Based on the Type A quota that was announced in China just yesterday, we believe we will start converting these orders to revenue and age in, although we do expect 3 to 6 months of lag time between the announcement of the quota and revenue conversion as the end user hospitals are required to go through the regulatory license application process for these devices.

We anticipate second quarter net age-outs to be approximately $31 million, of which 20% is represented by an additional amount of China Type A systems. This does not reflect the benefit of the quota announcement from yesterday as we are still in the process of evaluating such benefit.

Due to the higher level of age-outs, we ended first quarter with a backlog of $461.9 million, representing a decrease of 1% over prior year. We do expect our backlog to grow in Q2 and for the full year.

Turning now to our income statement. Total revenue for the first quarter was $95.8 million, representing a 5% increase over prior year. APAC and Japan were the primary drivers of our revenue growth. Product revenue for the quarter was $41.5 million, an increase of 7% over prior year. The product revenue increase was driven by strong demand for our Radixact System, which more than doubled year-over-year. Since its introduction, we have now recognized revenue for approximately 60 Radixact Systems.

Service revenue for the quarter was $54.3 million, an increase of 4% over prior year. The growth in service revenue was primarily driven by an increase in treatment planning and connectivity software price purchased through service contracts.

Turning now to gross margin. Our overall gross margin for the first quarter was 39.5% compared to 41.9% in the prior year. Product gross margin was 40.9% in the quarter compared to 43.2% in the prior year. The decrease in product gross margin was primarily driven by the product mix with a larger percentage of our overall sales attributable to the TomoTherapy platform compared to the prior year. We do continue to see a premium being paid for latest generation of Radixact Systems compared with our older generation TomoTherapy systems.

Service gross margin in the first quarter was 38.5% compared to 40.9% in the prior year. Prior year Q1 service margin represented the highest single quarter service margin in the company's history driven by low power consumption and other costs. Our fiscal 2018 service margin was 36.6% and our fourth quarter fiscal 2018 service margin was 37.4%. Therefore, the current quarter service margin of 38.5% represents continuous improvement in reliability and service efficiency.

Moving down to income statement. Operating expenses for the quarter were $42.6 million, an increase of 6% from $40.2 million in prior year. The first quarter G&A expense included a onetime $3.7 million accounts receivable impairment charge related to 1 particular customer. Excluding this onetime charge, the first quarter operating expenses were $38.9 million, representing a decrease of 3% over prior year.

On a sequential basis, operating expenses decreased approximately $2 million from $44.9 million in fourth fiscal quarter of 2018. Excluding the impact of the onetime impairment charge, operating expenses decreased $6 million sequentially. The sequential decrease in operating expenses was primarily due to 3 reasons: First, we had trade show related expenses in the prior quarter that did not recur in the current quarter; second, lower compensation costs related to the headcount management and commission associated with revenue levels; third, lower R&D cost as we near the release of motion synchronization for Radixact System.

First quarter operating loss was $4.7 million compared to the loss of $2.1 million from a year ago. Adjusted EBITDA for the first quarter was $4 million compared to $3.1 million in the prior year. First quarter adjusted EBITDA of $4 million excludes the impact of onetime $3.7 million impairment charge discussed earlier. We ended the first quarter with $71 million of cash and short-term restricted cash. The decrease in cash from the prior quarter was primarily driven by the payout of employee bonus early in the prior fiscal year as well as procurement on inventory in anticipation of fulfilling customer orders for the remainder of this fiscal year, both of which are seasonal in nature.

We expect to generate positive operating cash flow for the remainder of this fiscal year.

Before I move on to update our fiscal 2019 guidance, I'd like to discuss the targeted financial impact of the cost-reduction initiatives Josh discussed earlier. We anticipate taking a onetime charge of approximately $1.5 million to $2 million in the second quarter related to the reduction in workforce. The total annualized savings from this action is expected to be approximately $15 million, and we expect to start realizing the full benefit of this cost reduction in the fourth quarter of this fiscal year. Of the expected savings, approximately 30% will benefit gross margin, while the remainder will benefit operating expenses.

Turning now to our guidance for fiscal 2019. The following guidance for fiscal 2019 does not reflect the potential benefits of the China Ministry of Health announcement yesterday of Type A and B quota and license, as we are still in the process of analyzing such benefits. As Josh mentioned earlier, we will communicate the potential benefits of this development when our analysis is complete.

We are reaffirming the revenue guidance provided back in August, which is for annual revenue in the range of $415 million to $425 million representing growth of approximately 3% to 5% over fiscal 2018. The guidance includes 4% to 8% of product revenue growth. We continue to expect overall gross margin to essentially match fiscal 2018 gross margin of nearly 40% as we anticipate TomoTherapy, Radixact platform revenue to be a higher percentage of the total revenue in this fiscal year.

We are updating our adjusted EBITDA and operating expense guidance to reflect the impact of the cost-reduction initiatives as follows. We now expect our adjusted EBITDA to be in the range of $23 million to $29 million, excluding the impact of the $3.7 million impairment charge in the first quarter and the $1.5 million to $2 million severance charge in the second quarter. The updated EBITDA range would represent year-over-year growth between 35% and 70%. This is an increase from our previous EBITDA guidance range of $21 million to $27 million and reflects the direct impact of our cost-reduction initiatives.

As previously mentioned, the updated guidance reflects our expectation that the full benefit of cost initiatives, which is approximately $15 million on an annualized basis will be realized in full starting in the fourth fiscal quarter. We now expect operating expenses to be down approximately 1% year-over-year, excluding the impact of onetime impairment and severance charges discussed earlier.

And with that, I'd like to hand the call back to Josh.

Thanks, Shig. Before we open up the call for your questions, I'd like to thank the entire Accuray team for their continued focus, commitment and improving execution, supporting the important work that's making a difference in patients' lives. Operator, we're now ready to open the line for questions.

(Operator Instructions) Our first question comes from the line of Brooks O'Neil with Lake Street Capital Markets.

I understand that you are delaying comments on the China situation, but guys, could you just say whether the gross numbers are consistent with what you had expected previously?

Brooks, this is Josh. Yes, very definitely in line and consistent with our expectations. 188 Type A licenses, 1,208 Type B licenses. Again, very consistent with what we were expecting. And we're -- well, we're not going to be commenting in detail on what it translates into relative to potential benefit or positive impact. We're very -- obviously very pleased. This has been a long wait for us. And as you might imagine, we're pretty excited about it.

That's fantastic. I'm excited too. So hopefully that will begin the flow-through maybe even later this year. So that will be good. Secondly, I was hoping -- you understand I think that I'm not a clinician and hence, all the noise out there related to adaptive radiotherapy and the differences between MRI guidance and CT guidance admittedly generally go right through my ears and out from one side to the other. So I was hoping you might be able to describe your high-level feeling about your capabilities, particularly, as it relates to the motion synchronization with Radixact and CyberKnife relative to the capabilities that are being touted by some of your competitors in the field.

Yes. That's a mouthful kind of a question, but let me take a stab at it. So to start with Brooks, we have never said that we don't believe better image resolution is important. Quite frankly, we believe absolutely that it's important. What we've been saying from the very beginning is that, there is really a balance and a trade-off between improved image resolution on one hand and -- especially soft tissue resolution, and both product economics, which is the cost side of that equation and the functionality or the speed and the throughput, if you will, of the MR-based systems, because they kind of, by their very design, they are going to be more difficult devices to generate large-volume patient treatment schedules and support large-volume treatment schedules on a given daily basis. Our view in general is that as we -- if we look at our road map and we have made improvements already and are continuing to advance our imaging capabilities to kV cone beam CT capability, we believe, along with the Synchrony software technology in terms of our devices -- CyberKnife already and soon to be Radixact, our devices being able to track motion, correct for motion and ensure beam synchronization automatically in the form that we deliver it will allow for real-time adaptive therapy capability. And again, there is no question that everyone would like to see improved soft tissue resolution and imaging improvements. And we're -- in our view, we're advancing our pipeline in the ways that provides the best balance between, again, enhanced resolution, improved imaging and the other elements that I just described, the economics on one hand and the overall efficiency or throughput of the device platforms on the other.

That's very, very helpful. I just have one quick follow-on. You were talking about the efficiency, the throughput efficiency, and my sense, obviously, in talking with you and others is that your ability to adjust the beam in relation to what you see is quick and efficient, if I understand it correctly. Do I have that right?

Yes, yes. There's also, what we're doing is being done by the software algorithm and the hardware. And there's no -- essentially, there's no gating involved. There's no -- we're not making either adjustments to -- or the position of the patient treatment couch or turning beam on and off. So all of those elements, again, create just better throughput vis-à-vis efficiency.

Our next question comes from the line of Anthony Petrone with Jefferies.

Also congratulations on the announcement out of China last night. It's been a long road on that front. Maybe just, Josh, we can start there a bit, and again, realize that the information is going to be limited here. But when we look at the numbers 180 and 1,200, can you maybe crosswalk where Accuray left off? I think, if our memory and notes serve us correctly, Accuray already had about 39 Class A licenses secured maybe going back a year or so. I'm not sure the Class B licenses were ever disclosed, but maybe just a crosswalk on where Accuray stood in all of this?

Sure. So if you go back to when Type A radiotherapy licenses were actually still being issued, Anthony, 37 -- a total of 37 had actually been issued. They had authorized actually more than that. I think the number was somewhere in the 55 or 60 range. Only 37 of that 55 to 60 had actually been issued, and 34 of the 37 were awarded to Accuray devices either TomoTherapy at the time or CK. So I mean, that's going back now probably almost 3 years ago, it's a long way back clearly. And I don't think we're expecting that we would going forward operate at that level of win rate. I mean, I think the calculation on 34 out of 37 is something like 87% or 88%. That's not in our assumptions. But, quite frankly, if you look at how effectively we're in that space back then, we feel really quite frankly, really bullish about what the future holds with licenses now starting to flow again. I think the other thing I would just reinforce, we've talked about in the prepared remarks, is that, again, based on prior experience, which is really the only thing we have to go on, we would expect that you probably see a lag time between a quarter or 2 from, again, time of license announcement until the time that we start to see revenue impact related to those licenses starting to flow.

No, it's helpful. And then just to be clear, taking another derivative within there would be flow of gross bookings. So now that there is an increased number of licenses available to 2020, this would also presumably reflects in quarterly bookings going forward. I mean, how should we think about the bookings impact of all of this?

It's absolutely going to impact orders as well, without question. I mean, I think our distributor in China was -- as you might imagine, was looking as forward as we were to finally getting an announcement. And I think that we would expect that we'd start to see some flow -- incremental flow of orders from a here now that the market has clarity around license availability. Just to provide, Anthony, just a little bit more color real quick on -- just to kind of speak to the timeline or the lag from here to when we would start to see some activity, this process essentially starts now with the Ministry of Health establishing the number of licenses for both Class A and Class B radiotherapy. Hospitals, which -- who are looking to procure or purchase devices in either Class A or Class B are, in essence, going through an application process, submitting that application process or submitting their application, which, in essence, will create somewhat of a bidding situation for the defined number of licenses that have been issued or that will be issued and the number of accounts or hospitals that are interested in accessing those. The criteria that will be used there will be geographic location, where in the country they are, are they ready to accept devices in the Type A space that are at a higher level of specification from a facility capability standpoint, a staff training standpoint, those kind of criteria. And then once granted, they would receive an import license that would be granted to the facility itself that would allow them to go forward with the purchase. So that kind of just ties together that 1 to 2-quarter lag time that Shig talked about in his prepared remarks.

That's great. And I'll do one more follow-up just to go -- maybe on the CyberKnife side, and I'll get back in queue. We had good feedback on VoLO. I'm just wondering can you kind of describe what VoLO does to CyberKnife treatment times? And what you think VoLO could do for CyberKnife's demand going forward?

Sure. So no one has ever debated or challenged the validity of the precision and accuracy of CyberKnife. What has been a challenge for people, certainly in the earlier generations of the device, was the amount of time it took to develop a treatment plan and to deliver the treatment. And if you look at where we've come from over time, both with the latest generation platform, which is the M6, the introduction of the MLC, multileaf collimator, for the M6, we've seen significant improvement over time. The VoLO optimizer really kind of take this to a different level. It's going to reduce treatment planning time, so the time it takes to develop a treatment plan, by as much as 90%. And it's going to reduce the amount of time that it takes to deliver that treatment by almost 50%. Now that may vary across -- the time reductions might vary across the type of cancer, prostate versus lung versus spine. But in general, I'd say those numbers are reasonably accurate in terms of what we're seeing and what users of the optimizer are telling us with regards to their experience. So it takes -- quite frankly, in an example like prostate, with organ-contained cancer, it will reduce the treatment time to something that probably is sub-15 minutes, maybe 12 to 15 minutes. And that's really, again, from market experience and clinical experience with CyberKnife, that's an extraordinary leap from where we have been historically. So our optimism is high that the introduction of the VoLO optimizer allows us to have a catalyst, or creates a catalyst, if you will, for trade-in, trade-up opportunities in terms of the installed base and the replacement sale opportunity, which is primarily by geographic concentration, primarily a U.S. opportunity.

Our next question comes from the line of Amit Hazan with Citi.

This is [Phil] on for Amit. Can you hear me okay?

Yes.

Great, great. I want to follow on a little bit to Anthony's question on China. Just to make sure I'm clear on the lines there between Type A and Type B. I think we've talked previously about the strategy with Onrad and TomoH. Can you just clarify sort of where the lines are in terms of the product decisions between Type A and Type B licenses?

So the Type A radiotherapy category by definition are the higher-end products, higher end both defined by price point and by technical specification. I think you'd -- maybe not exclusively, but I think you would tend to see more of a Type A radiotherapy products in academic medical settings, maybe in the large PLA military channel hospitals and probably the larger public hospitals in the market. Type B are products that we're calling essentially the value segment of the market, capable, efficient, but probably more in line with devices that you'd see in either freestanding treatment centers or smaller facilities, doesn't mean that they're not effective and not good products just, again, different product for a different segment of the market. But if you look at the Tier 1 hospitals, the largest hospitals, you're talking about products that -- essentially requirements that would tend to lend themselves or line themselves up with Type A radiotherapy licenses. The big volume, the unit volume opportunity here going forward is in Type B. And also, really primarily and probably more of the private sector. If you look at how the market will develop and what the expectations are for growth going forward.

Okay. That's very helpful. I think, is there any additional clarity you can give on that break-even price point? Is it a clear line? Or is it sort of a confluence of the specifications versus the price paid? Is there a throughput consideration? Anything along those lines?

Not that I'm aware of at this point. We can try and get back to you if there is something that either emerges or something that is beyond or incremental to what we currently know. But I think in general, what you'd see is products from a price point standpoint in Type A, products that are north of -- at the end user level, certainly north of probably $2.5 million or $3 million.

Okay. That's very helpful. Just one more quick one if I can follow on, on the financial side. Is there any clarity you could provide around the embedded foreign exchange or currency, either on your top line or maybe on top line and on OpEx as well for the quarter and for the year? And any color there?

Yes. So our growth for the product revenue, just to kind of you give a sense, was 7%, but, [Phil] constant currency, that one was 8%. So very minimal 1% uplift. Service was a little larger, 4% growth we talked about, on a constant currency basis by 2%. So it would have been 6%.

Okay. So we saw an uplift in the first quarter. Are you expecting essentially mutual currency for the year with the reversal of currency?

That's kind of what we're expecting.

Our next question comes from the line of Brandon Henry with RBC Capital Markets.

Can you spend some time discussing the various areas that are contributing to the cost savings? And how should we think about the ramp of cost savings over the course of fiscal year '19? And then just on that latter point, can you also help me reconcile the $2 million increase in the adjusted EBITDA guidance with the $15 million in cost savings? And I just got a couple of follow-ups.

Sure, Brandon. So the $15 million of the annualized savings, as I mentioned in my prepared remarks, 30%, we expect to benefit the cost of goods lines of margin and the remainder to be in the OpEx section. And when we think about the functional areas that were affected, it's mostly support the G&A-type areas. Josh and -- we were very careful about not impacting the ability to execute commercially. So the areas we cut were not related to commercial operations. As far as how to kind ramp the savings up next 2 quarters, Q2, Q3, we should see some savings, but these effects were -- are being offset by a couple of things: Number one, it does take some time to realize the savings on a personnel side of things due to transition that you can expect in personnel. And also, there are some expectation that we would spend a little bit more on a JV -- China JV-type projects, so it's advisory, legal, those type of costs. And so that's why we don't expect to have a full benefit until Q4. And I guess, to your last point, so $15 million of annual savings, and I'm saying the full benefit to be in the fourth quarter. So let's take 1 quarter, so it's about $3.5 million, 70% of that goes to OpEx in Q4. If you think about Q4 exit rate last year, we were about $45 million. So I think if you want to try to take down a couple of million dollars from that rate $45 million in the last year Q4 to project out to a Q4 exit rate this year. I think that's what you're looking at.

Okay, very clear. And then, on the China JV, can you just give an update on where you're at with that? When we could expect a deal to be finalized? And then how long does that really take to get fully up and running?

So Brandon, the work that we're doing -- we are in exclusive negotiations, we got an executed term sheet and we're in exclusive negotiations with a final party, if you will, or a counterparty. We've said, we're working hard and fast at this as they are. So they are as excited as we are to get going or to get to at least and execute a definitive agreement, but difficult to say. I mean, obviously, this is an important part of our strategic growth agenda going forward, given the opportunity in China. We want to make sure we get it right. And I think there's a hope and aspiration that this is something that we could potentially get to around the end of the calendar year, but again, impossible to predict timing on this. So it's more important that we get it right then we -- and complete it in a fashion and with the elements intact that are right for both parties. We expect it to be a long marriage. And getting it right is more important than getting it done quickly. Although again, there's a lot of energy and a lot of focus on both sides related to do that. On the question related to time line or the ramp, the longest part of the ramp for the JV will be the activities related to the development of a product, a Type B product, which will essentially be a product that would be manufactured on the ground there by the partner. They'd have responsibility for the regulatory registration and approval process for the device. And it would essentially be considered in the eyes of the market, really a true local domestic ethnic Chinese brand. So when you think about starting up a manufacturing facility, the registration process, we're probably talking about something that's, I guess, 18 to 24 months away in its time line.

All right. And then just last question from me. Can you provide any more details on the AR impairment charge in the quarter?

Yes, Brandon. So it is truly is a unique case that we had. It's a standalone cancer center, 1 customer who just bought once from us a couple of years ago. Unfortunately, the cancer center never took off. So hence, we took a charge this quarter. We don't have any further exposure to this customer at this point.

Our next question comes from the line of Sean Lavin with BTIG.

This is Marie Thibault on for Sean. I wanted to start with the net orders this quarter. So the difference between gross and net was a little larger than I expected, and I think a little larger in terms of age-outs than had been guided. I understand a lot of that had to do with the Type A orders in China, but I wondered if there is anything else in terms of cancellations or age-outs that you wanted to call out in the quarter? And I think you also mentioned other adjustments. I'm curious with that.

Yes. Thank you, Marie. So yes, I pointed out the China being the large part of that, $26.5 million age-outs, as we said we hope to be able to promote the revenue with announcement that we heard. I think the other aspect we also point out that distribution revenue on orders continue to be a high percentage in the recent years, so I think we are about 70% range exit in the Q1. So again, we continue to focus on converting those -- distributing orders to revenue, but in some areas do take some time. I think from cancellation perspective or from a quarter-to-quarter, that can vary, but I think for the annual basis, we're comfortable that, that still remain in the low single digit in relation to our total backlog. The other adjustment -- the small onetime adjustment related to our 606 transition, I will point that it's less than 1% adjustment of the backlog and the small FX impact in that number.

Okay, great. And I wanted to hear a little bit about multi-systems and how the order funnel looks there? I know that's a big part of kind of the U.S. growth story, so I'd love to hear anything anecdotal or anything that you can give us on that front.

So I think, as we've talked about in the previous quarter, we have an improving funnel -- or a level of visibility in the funnel in the Americas, and primarily U.S. that looks like there are more multiunit opportunities for both CK and Radixact than we've had in the past. Again, because of the complexity, in general, with those -- that selling -- those selling processes or situations, Marie, more different facilities involved, more decision makers or influencers involved, so in essence, more touch points from the customer side, it becomes a little bit more difficult to more -- to hone in or more accurately predict timing of some of those. But in opportunity set, we feel good about an improving situation in terms of those opportunities. I should point out, though, that the multiunit orders situation is not the only driver in the U.S. going forward. We had -- in the quarter, we talked about how Radixact was a contributor -- how big of a contributor Radixact was in terms of gross orders. I think the other thing that's occurring in both the U.S. and in Western Europe from a direct sales standpoint is the customer profile is expanding. Our traditional strengths have been in larger academic medical centers, research-based medical centers with larger footprints in terms of radiation oncology department, number of vaults, et cetera. Over 70% of the first quarter Tomo activity or Radixact activity was in single- and dual-vault settings, which is a pretty strong indication that we're operating in smaller, either community or regional-based hospitals to a greater extent than we have in the past. And I think that's another reason for optimism in terms of longer term our U.S. situation and the contribution we can expect from the U.S. business to be playing at a bigger level.

That's great color. I appreciate that. One last quick one. Just as we kind of think ahead with China and the licenses, can you remind us or tell us what kind of a -- is there any, I guess, abnormality in terms of kind of a book-to-bill time turnaround in China? Is it longer than typical cycle, shorter than typical cycle? Anything you can give us there?

I think if we go back to where, again, the period of time when we were very actively taking orders and winning orders in the Type A segment, I think what -- we were probably operating to shorter time in the backlog during that period. In a directional sense, I'd say probably shorter back then.

(Operator Instructions) Our next question comes from the line of Josh Jennings with Cowen.

I had to hop on another call, so I missed some of the prepared remarks and some of the questions, so feel free to say that you have addressed these questions if I'm making you repeat yourselves. But maybe to start, just to talk, again, I was going to ask another question with JV in China opportunity and then, maybe -- I mean how would you have us think about the opportunity there for Accuray, just from a size perspective, and I think all your competitors talk about emerging markets and China specifically as an opportunity, but having a domestic, I guess, footprint on the ground, maybe you could just update on your thoughts on what that China opportunity can actually look like once you're up and running on the manufacturing front?

So Josh, we've said pretty consistently over time that the big opportunity from a growth perspective over time and in terms of unit volume, is really in the value segment, which is the Type B -- would be the Type B specification or Type B category products. We have Onrad, we have H Series, Tomo. But quite frankly, we don't think that portfolio is enough longer term. And there are also, we think, very distinct advantages to be operating in more of, again, from a positioning perspective as more of a domestic ethnic brand. And when we started to do the calculus around how do we play, go-to-market strategy, et cetera, we pretty quickly came to the conclusion the risks to a company of our size of trying to go that alone, down that path alone was really -- it was just not a calculus and risk profile that we felt comfortable with. Finding someone from a partnership standpoint that has a strong market presence, and this partner does, although not necessarily today in the radiotherapy space, but they're in the health care market product space in a big way, they're a market leader in their core business. They have strong market access, a large number of existing accounts throughout the country. So it's very much a business that has infrastructure reach into the provinces, which we think is also where, again, growth comes from longer term. So for all of those reasons, quite frankly, we think that's -- this is the better play for us in terms of strategy. We get to partner with someone that provides great leverage for us and our products in the core technologies and gives us a chance to kind of mitigate some of the risks that we'd be taking on if we're trying to do all of this by ourselves. So that's -- I mean in great part, that's how we're thinking about the JV and longer-term participation in a market that we believe is going to outpace growth, quite frankly, virtually everywhere else in the world.

I mean, would you say that this is a $500 million market opportunity, $1 billion market opportunity? I mean, I guess, maybe just to talk about the Class B opportunity in China, and then how much growth you see over the next 5 to 10 years within that segment?

So if you just think about the sheer number of hospitals in the country, and not just the existing ones, but the ones -- the projected growth rates of new facilities. I mean, there are new facilities being built private sector funded, private capital funded, probably not a lot more or not nearly the kind of growth on a public or PLA side that you'll see comparatively with what the growth would look like prospect-wise on the private side. There's, again, I mean, the estimates are that you're talking about in the aggregate for all product types, a market that, again, it depends on whose numbers you like today, but something that would be maybe on the conservative end of the spectrum, 5,000 incremental devices. If you really wanted to get very aggressive, go 100% more than that, say 10,000. But there is a vast, vast under capacity today in radiotherapy capability in the ground available to treat patients today. And you start to do the math around the demographics. The estimates are 15 years from now, Josh, there'll be about 0.25 billion people, age 65 years or older in China and you overlay that population, the growth in that aging population overlay it on top of what, again, the projections would say are some of the most aggressive projected incidences and forecasted rates of cancer diagnosis by anywhere in the world, especially in lung and liver. So maybe it takes some of that -- some of those data points have kind of work your own math on it, but it's -- I'm not sure that anybody can actually calibrate how high up is on this, but it's a big number.

That's helpful. And then, my last question and again, please free to tell me that you've already addressed it. But historically, we've talked to radiation oncologists about the CyberKnife and one of the issues or hurdles has been treatment times and you did kind of launched this VoLO optimizer where planning and treatment times have come down significantly. You have the prostate data specifically but also lung data specific to the CyberKnife as well. But can you talk just about the reduction in planning and treatment times that you're going to go forward with, with the CyberKnife? And how meaningful that can be in terms of claiming one of the adoption hurdles of prolonged treatment times?

Yes, so I'll give you the abbreviated response because we did spend some time on that in an earlier question. But we're talking about an impact to treatment planning time that would be -- the time to develop a treatment plan that would be a reduction of almost 90%. So I mean, it's sea change kind of difference. And in treatment delivery time, again, you probably have some variation in terms of the location or the type of cancer. But treatment delivery time reductions of roughly 50% or higher depending on where your -- what kind of treatment you're talking about. If you just use prostate as kind of a proxy example, we think we're going to be able to get with essentially a CyberKnife M6, which is the latest generation equipped with a multi-leaf collimator with CyberKnife with the VoLO optimizer, sub-15 minute treatment time on a prostate case, probably something like 10 to 12 minutes, which would be, again, dramatically different than what the market has experienced over time. We've been making incremental improvements over the last 4 or 5 years, the M6 generation device, first impact, the MLC, the next impact. But the optimizer really, really truly makes this a different product going forward. And one that we think is a catalyst for the trade-in, trade-up activity that exists or opportunity that's out there in our installed base with the older devices that are coming into replacement.

Our next question comes from the line of Tycho Peterson with JPMorgan.

Josh, I actually wanted to follow up on your last point there on upgrading the installed fleet of CyberKnife systems. You got the optimizer, but then obviously you got the kVCT imaging coming near further down the road. How much of the catalyst do you think VoLO is to upgrade customers? Or do you think they potentially wait until you put firmer timelines on the kV upgrade?

Based on feedback -- early feedback from people that have used optimizer and the feedback we got last week, I would say I know that there are conversations taking place already from a trade-in, trade-up standpoint. The optimizer is a big -- I think it's going to be a big catalyst for us.

Okay. And then, I guess, shifting over to China and I appreciate all the color. I'm just trying to reconcile. You got the commentary on that Type A and B quotas but obviously, you're still waiting to announce the JV partnership. I guess, how much of that quota do you think you can penetrate before you actually have the partnership up and running and actually have some new lower cost equipment coming out of the back of that?

Again, I mean, we said based on our experience from prior times, back about 3-plus years ago with -- when Type A licenses were still essentially being issued in their earliest phase, we were seeing from the time licenses became available to the time we're starting to impact revenue and take revenue, somewhere between 1 quarter and 2 quarters. So you would think that we would start to see some impact this year of this announcement from last night. I think that, again, we're not going to get specific about how much impact. I think that's on the revenue side of the discussion. I also think it's -- I answered an earlier question, I think it was with Anthony Petrone from Jefferies. I think the order impact is going to be significant, too. I mean, our distributor, as you might imagine, has been waiting as earnestly as we have for this announcement in their -- they are -- I'm pretty certain they're going to hit the ground running in terms of selling opportunity here. So again, that's what -- again, what we know based on what our prior experience has been, the unknown on the other side of the equation based on your question is how long it takes us to get to a completed agreement, a definitive agreement on the JV side. And again, as I mentioned earlier, I don't think that we can put a stake in the ground on how long that's going to take. It has our full attention. It has the full attention of our exclusive counterparty that we're negotiating with, and I think it needs to be thought of as kind of a stay-tuned discussion vis-à-vis timing.

And then in your prepared comments, you did allude to some pricing sensitivity around the tariff dynamic in China, are you able to kind of elaborate on that at all? And does the risk there go up with this third round of tariffs?

Yes. On an import basis, it was material on the import side. From a China to U.S., I mean it's probably $1.5 million annually if that -- we don't manufacture full systems in China. We manufacture a linac, linac head, linac waveguide in Chengdu. And we've got a flat panel detector that we purchase from a supplier in China. But -- so it's minimal from an import standpoint. From an export perspective in terms of devices sold into China, our distributor contracts actually specify today that the distributor carries the burden of the tariff. I would tell you that maybe some amount of pricing negotiation took place for Q1 orders. But based on very healthy high user -- end-user pricing in China, I think, we're comfortable in saying that we've got line of sight of continuing healthy margins in that part of it.

And then just lastly on Radixact, can you maybe help put in context the ASTRO announcements, the CTrue Iterative Reconstruction and then real-time motion synch, which is obviously pending, how important do you think those are to kind of driving adoption, obviously Radixact orders have been strong anyway, but I'm just curious what type of the feedback you've got?

I think they're very, very important. We've been talking about our road map and our upgrade pathway for Radixact for some time. And the fact that we have motion and essentially the Synchrony, the Synchrony software and that technology submitted from a 510(k) standpoint is really meaningful. We said that we have -- we estimate based on kind of projected turnaround times you can never guarantee them, but we think it's a relatively straightforward regulatory submission. And based on that, we could have reference site and ramp and monitor site installation and patient treatments taking place before the end of this calendar -- sorry, before the end of this fiscal year, which would be a terrific outcome from a timing standpoint. And we've got a large number of Radixact-based systems already installed or orders for them in the backlog that were -- the premise for those customers placing orders was their interest in the upgrade cycle, the innovation pathway that we talked about from a pipeline standpoint. So I think these things in the aggregate are pretty material contributors to a growing level of interest in the product. Again, we took -- it was a strong quarter in general for Radixact. And the interesting thing, again, I mentioned before is that the customer types, these were not just all academic medical centers. 70% of the orders that we took for Radixact in Q1 went into single- and dual-vault locations, which meant in general, by definition, you're talking about smaller facilities, maybe community or regional-based hospitals or even freestanding centers.

There are no further questions at this time. I would now like to turn the call back to Josh Levine for closing remarks.

Thanks, operator. And thank you, everyone, for your participation this afternoon. We look forward to speaking with you on our fiscal second quarter earnings call. Thanks, again.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude today's program. You may all disconnect. Everyone, have a great day.