Invitae Corp (NVTA) 2017 Q3 法說會逐字稿

Good afternoon, my name is Chris, and I will be your conference operator today. At this time, I would like to welcome everyone to the Invitae's Third Quarter 2017 Financial Results Conference Call. (Operator Instructions)

Kate McNeil, Head of Investor Relations, you may begin your conference.

Thank you, operator, and good afternoon, everyone. Thank you for joining us for our third quarter 2017 earnings call. Joining us today are Sean George, our CEO; Shelly Guyer, our CFO; Lee Bendekgey, our COO; Dr. Bob Nussbaum, our CMO; and Katherine Stueland, our Chief Commercial Officer.

As you listen to today's conference call, we encourage you to have our press release available, which includes our financial results as well as metrics and commentary on the quarter.

Before we begin, I'd like to remind you that various remarks that we make on this call that are not historical, including those about our future financial and operating results, our plans and prospects, the focus of our business strategy, our plans to integrate and manage businesses we acquire, our expectations regarding Good Start's business and financial results, our proposed acquisition of CombiMatrix, market opportunities, future products, services, our product pipeline and the timing thereof, demand for and reimbursement of our services and our investment in our infrastructure and operations, constitutes forward-looking statements within the meaning of the safe harbor provision of the Private Securities Litigation Reform Act. These forward-looking statements involve significant risks and uncertainties that could cause our actual results to differ materially from our guidance. Our guidance on future company performance assumes, among other things, that we don't conclude any additional business acquisitions, investments, restructuring or legal settlements. We refer you to our 10-Q for the quarter ended June 30, 2017, in particular to the section titled Risk Factors, for additional information on factors that could actual results to differ materially from our current expectation. These forward-looking statements speak only as of the date hereof.

With that, I turn the call over to Sean.

Thank you, Kate, and thank you again, everyone, for joining us on the call today. On the heels of a strong second quarter, with increased confidence, we made swift moves to expand our reach into family health, namely with the acquisition of Good Start Genetics in August of this year and signing the definitive agreement to acquire CombiMatrix, which we expect to close sometime in November. And as we exit Q3, I'm incredibly proud of this team's focus and execution during what turned out to be a very eventful quarter. We did experience a bit of turbulence along the way, some of it expected, some of it unexpected, namely, the beginning of integration activities for Good Start Genetics; the Anthem implementation and rollout of AIM; of course, CMS pricing in advance of the PAMA rollout; and then an issue in production with the Boland inversion. Throughout it all, the team demonstrated focus, swift execution. And while managing a lot of moving parts, we came out of the quarter with increased momentum as we head to the rounding out the end of the year. I'd like to give some management perspective on each of these issues before handing the call over to Shelly to cover the numbers in more depth.

Starting with the challenges we anticipated. Of course, the beginning of the integration for Good Start, with our primary prioritization being -- focused on the commercial integration. In particular, sales leadership was engaged in integrating the commercial efforts so that we can enter the year -- next year with one combined sales force offering all of our services. And we even added some of CombiMatrix services through a strategic distribution agreement. The whole organization is aligned and rapidly integrated. We've combined the day-to-day operations and systems. And indeed, just a short 2 months after closing, we've got full integration of both operating teams.

Now as can be expected with -- often arises in acquisitions, sometimes the third-party relationships -- pre-existing third-party relationships get frayed. And indeed, the former Good Start relationship with Sema4 had been terminated. This happened fairly early in the process. It was the termination of the send-out agreement for expanded carrier services. So that was something that, again, while not totally expected, is something that often comes up when you do these things. Took some time to manage the clients and, of course, we'll accelerate our development to bring up an Invitae version of that expanded carrier screen as soon as possible.

So while we sit today, the integration is going about as well as one would hope. There's still work to do in the following 6 to 9 months, particularly in preparing for the transfer of production operations to San Francisco.

A couple of things in the quarter that were a little more challenging than we anticipated. I mentioned Anthem launched the AIM portal. It's an automated prior authorization portal. That happened in July. It did, indeed, cause a short-term disruption, primarily because of some confusion on the client's side in the selection, order and use of the portal. It was, frankly, a little bit of a rocky rollout. We've been working with the clinical practices, starting at the top, working through the issues. We've got a good relationship, of course, with Anthem itself, working on issues, both in the presentation and implementation of the tests in the portal. Overall, it was about a $400,000 revenue impact in Q3 of '17.

Now in the long term, the -- both this and other efforts by payers in the long term are actually net positives. A standardized enforcement of codes, the use of codes, demonstrably increases the predictability of coverage, and in particular, and perhaps most importantly, increases the utilization of in-network providers. Therefore, as an in-network provider for the largest health care insurers, provides us with an advantage versus labs out-of-network attempting to get volume into clients.

Now some good news in all of this is that early indications are that United's Beacon platform, a similar platform, is indeed going a little bit smoother. United had launched a version of that in a pilot basis in Florida, had spent a little more time with webinars and client education on how to use the portal. And while it's early, early returns are in, it would appear that it's going a lot more smoothly than Anthem.

And then, of course, late in the quarter, CMS released the newly proposed pricing to be rolled out with PAMA next year. And now as we had indicated a while ago, our collective experience with CMS upon new pricing implementation is 3 -- is kind of always 3 things: it's always confusing; typically changes over the months to come; and usually, while it doesn't start this way, it usually ends up in a fairly rational point that makes sense the most.

So while we're expecting a bit of noise from this, from both the announcement and dealing with it, just how confusing and seemingly misconstrued the results were surprised even us. We have commented in the official public commentary period. More importantly, we, of course, are engaged in all manner of conversation through all channels to help clarify what we think was a misconstrued calculation of the pricing for the code that we've been instructed to use as well as overall some confusion about the use of codes more widely.

It is our understanding that PAMA is on track to be implemented in January 2018. And so I'll remind people that the worse case, if everything remains as it is today, that Medicare accounts for about 8% to 10% of our revenue. And the current proposed pricing, again, if nothing changes between now and then, is a 10% drop on that revenue in 2018.

Now one of the biggest challenges that I mentioned in the quarter, and certainly, the one that we were not expecting, the Boland inversion issue was brought to our attention. We immediately jumped on it; got in touch with all of our clients; communicated completely, openly, transparently; proactively engaged on understanding more; finding out what the issue was; maintained constant communication as we investigated it ourselves and started to clarify what the shape and form of the remediation would look like.

Good news -- well, I'd say the good news is, it was about as we expected. And I think also, as we stand today, we are essentially done with the follow-up on that issue. We began notifying clients with changed results last week and are indeed finished with that.

We've identified over 50,000 negatives that we did confirm. There were 8 patients that had received a false negative result from us, so 8 positives that we identified. Again, those clients have been followed up with. And there are a handful of new samples that we're reaching out to collect just to double check. This was in line with our initial expectations, including the rough COGS impact. We mentioned it would be non-material. And indeed, the COGS impact for the quarter came in at less than $100,000.

Overall, the reaction from the clients is positive. And any volume headwind that we could try to calculate was pretty much in the noise of both Q3 seasonality and everything else I mentioned -- everything else going on in the quarter, that -- much of which I covered. Most importantly, our clients saw firsthand our commitment to the patients, the quality and integrity of our data. So building on the momentum that we had entering the quarter, and against the backdrop of all of the issues I just went through, this team here in Invitae, I feel, demonstrated the focus and proficiency of execution that over time, will earn us the right to emerge as a leader in the new age of genetic medicine.

So with that, I'll turn the call over to Shelly to walk through the detailed financials for the quarter.

Thank you, Sean. Before I dive into the details of the quarter, I'd like to remind everyone that our acquisition of Good Start Genetics closed on August 4. Because our consolidated results include only 8 weeks of contribution from Good Start, we believe that for this quarter, it is most informative to provide the consolidated financial results for the quarter, in addition to breaking out our base business. Additionally, we'll provide some color on what we believe Good Start's full quarter results would have looked like so our investors and analysts can better understand how the integration of this business and is expected to contribute to Invitae's results in the future. We do not intend to do such a break out regularly in the future, but we thought it would be useful in this quarter given the recent acquisition.

With that as background, let's take a closer look at the financial results for the quarter. As Sean indicated, it was a strong quarter, and our base business performed well. We continued to see solid growth in volume. On a consolidated basis, including Good Start volume, we accessioned over 40,000 samples in the quarter. This reflects not only year-over-year growth of approximately 188%, but also our 18th consecutive quarter of double-digit growth.

Looking at our base business, you can see Invitae accessioned more than 34,400 samples, reflecting 122% improvement over the prior year and strong double-digit improvement over the second quarter. Importantly, volume growth was seen across all clinical areas and is attributed to growth in new and existing accounts. Ongoing biopharma partnerships continued to drive our non-oncology test volume.

Good Start contributed approximately 5,600 samples since August 4. For the full quarter, sample volume was approximately 9,250 compared to nearly 9,500 samples accessioned in the second quarter. The biggest factor impacting the Good Start volume in third quarter was Sema4's termination of an agreement to provide expanded carrier screening services. At the time of acquisition, expanded carrier screen testing, or ECS, accounted for approximately 15% to 20% of Good Start's volume. While a number of clients that previously relied on Good Start's ECS offerings have adjusted their ordering habits, we expect that until we are able to introduce an ECS offering, Good Start's overall volume will remain depressed.

Revenue for the quarter came in at approximately $18.1 million on a consolidated basis, a 189% increase over the $6.3 million reported during the prior year period. Invitae's base business contributed approximately $16 million, reflecting 122% increase over the prior year period and a 12% increase over the second quarter.

Good Start's contribution to the third quarter revenue is a bit more complicated. For the 8 weeks following acquisition, we recognized approximately $2.1 million in revenue. In addition to the $2.1 million, Invitae received approximately $900,000 of cash related to tests billed prior to the close of the acquisition. However, U.S. GAAP does not allow us to recognize revenue on these cash collections related to tests reported prior to the close. As a result, this $900,000 increases the accounts receivable at the time of acquisition and does not count toward our revenue in the quarter or any future point.

Looking ahead, we expect to collect an additional $1.1 million from cash-based customers for tests conducted prior to August 4, with approximately 75% of these remaining collections occurring during the fourth quarter and the rest occurring in the first quarter or 2 of 2018. So we anticipate a similar impact on revenues on our fourth quarter results, and to a lesser extent, our first 2 quarters of 2018. For reference purposes, excluding the impact of acquisition accounting treatment, Good Start revenue for the full quarter would have been approximately $4.9 million, down about 6% from the second quarter revenue of approximately $5.2 million.

Moving to collections. We continue to work through the process of operationalizing a significant number of large payer contracts signed late last year. While we see progress on this front, improvements were offset a bit by Anthem's AIM implementation, which was rolled out in July. As Sean indicated, this is a long-term net positive for us. But it did have a negative impact of about $400,000 on the quarter.

Finally, you will recall that we brought billing in-house in June. The result of this effort has been impressive. Within the first several months, we have already achieved the operational efficiency of the third-party vendor operating for over 3 years. Over the few next quarters, we expect to see our billing team further improve collection rates and collection amounts.

Strong third quarter volume, coupled with some production improvements, enabled the continued reduction in cost of goods sold per sample. COGS for consolidated operations, including Good Start as well as for Invitae's base business, came in at approximately $330 per sample, representing an improvement of 27% year-over-year and about 4% from last quarter.

Looking longer term, we have opportunities to further reduce COGS and realize significant synergies in the consolidated business. In the near term, we expect to improve COGS through the full integration of Good Start, including the transfer of production operations next year.

On a consolidated basis, our gross profit increased from negative $1 million in the third quarter of 2016 to nearly $5 million in the third quarter of 2017. This includes a gross profit of $4.7 million from Invitae's base business. Our consolidated gross margin was 23%, and our Invitae base business achieved a 29% gross margin.

It's worth noting that our consolidated gross profit and gross margin calculations exclude the $900,000 of Good Start revenue that was classified as an increase in accounts receivable at the time of acquisition, and not revenue. For reference purposes, consolidated gross margin would have been approximately 30% had all Good Start revenues been recognized. I would reiterate that we remain confident in our ability to reach our long-term target of 50% gross margins across the Invitae platform.

We incurred operating expenses, excluding the cost of tests reported delivered, of $35.9 million of which about 32% were research and development expenses, 37% were selling and marketing expenses, and 31% were general and administrative expenses. In the third quarter of 2017, improvements in operating leverage were offset by nearly $2 million in acquisition costs.

Despite these higher costs, our leverage continued to improve. We drove 158% year-over-year volume growth while limiting increases in operating expense to 50%. For the third quarter, operating expenses included approximately $9 million in noncash expenses, including $4.8 million in stock-based compensation expense. Of note, this included approximately $1.3 million associated with the Good Start acquisition. Noncash expenses also included depreciation and amortization of equipment equaling $2.5 million. And due to the 3 acquisitions completed this year, our amortization of intangibles was approximately $550,000 in the quarter. As this only included a partial quarter for Good Start, we expect this rate to increase to approximately $700,000 in the fourth quarter.

Two final points should be made to understand this quarter's P&L. First, there was an income tax benefit due to the Good Start acquisition. Concurrent with the acquisition, the company recorded additional goodwill of $4.8 million relating to the tax consequence of recognizing the fair value of the acquisition-related intangibles, with an equal offset to deferred tax liability. And second, the weighted average shares used in computing our net loss per share does not include the non-voting preferred shares due to their dilutive effect. Thus, the shares used equaled 48.2 million shares principally related to the additional common shares issued during the quarter for the PIPE financing. The exclusion of the preferred shares had an approximate $0.03 negative impact on the net loss for the quarter.

We ended the quarter with approximately $106.5 million in cash, cash equivalents and marketable securities. This includes net proceeds of approximately $68.8 million from the PIPE financing completed in August and reflects the extinguishment of approximately $18.4 million in Good Start debt in conjunction with the closing of that acquisition.

As anticipated, we continued to see the cash burn related to our base business decrease. Our burn this quarter was $22.6 million on a consolidated basis. The burn for the base business was $20.1 million, a $1 million decrease from the second quarter. In the short term, we anticipate acquisition and integration-related expenses to increase burn. Longer term, we believe that the acquisition of Good Start and proposed acquisition of CombiMatrix will contribute positively to cash flow, allowing us to reach cash flow breakeven by the end of 2018.

Sean?

Thank you, Shelly. So as we can all see, the business is a bit more complicated than it used to be. But fundamentally, we are achieving our business objectives. I believe we're extending our leadership, both cost leadership and execution leadership in the industry, and bringing technology improvements into play in order to get genetic information in the hands of everyone and modernize health care systems.

At a very busy time, we not only continued to extend our growth on the core business and execute against it, but we added new assets, new capabilities and increased our strength. And in fact, I believe we took a quantum leap in our operational excellence, both on the operations side and our commercial execution. And every day, I'm reminded of our growing strength, whether it be from new biopharma partners seeking us out to work with us for our genetic testing expertise; or our network of clients, both clinicians and patients. Every time, we check in with the field or visit a key client, I'm reminded with new daily high sale volume that seem to come in on a biweekly basis. As you can kind of look and see the growing scale of our production operations and increasing leverage from our investment into our technology stack, you walk away with a clear distinction -- I walk away with a clear distinction of our future leadership potential. And indeed, as we kick off Q4 with October volume just on our base business, not including the prior businesses, coming in at nearly 15,000 samples accessioned, I believe it positions us well to achieve our 2017 full year guidance, and most importantly, set us up for a very exciting 2018.

And with that, operator, we'll turn the call over to Q&A.

(Operator Instructions) Your first question comes from Doug Schenkel from Cowen.

I just want to start on pre-authorization. You quantified the Anthem headwind. You're in network with Anthem. So I just wanted to try to get a better understanding as to whether these were delayed or lost revenue. So really, the questions are, first, were these tests done and not reimbursed? Or does this actually hold back volume as well as revenues? So that's the first question. The second question is, what was the rationale for the holdback? Were these tests deemed to not be within coverage policy? The third part is, what percentage of Anthem volumes were affected? And the fourth part is, what is the assumption in guidance before Anthem? Do you expect these tests to come back in the revenue? And for that matter, what are you thinking when it comes to United?

Right. Okay, so we'll do the four-parter. I can start -- clearly, I can start with the last. I think that going forward, we would expect to have worked this out. We have been, of course, on the phone with Anthem for the last couple of weeks. And I think, going forward, this should be immaterial. If it continues, of course, we will let people know. But it would appear that going forward, we've sorted this all out. The nature of it was not so much a volume impact, so I would say kind delayed is the answer to the first question. We do expect that revenue to come in. It was universally across the platform, so it's not as if the providers were sending one place versus another based on the rollout. It was just generally confusing. Test selection was confused. And in particular the nuances surrounding our specific agreement with Anthem and our menu offered, combined with what codes they wanted us to use and how it showed up in the portal, that was all very confusing. And some tests were ordered that -- for which clinicians were told weren't in acceptance guidelines, when in fact, we had worked those out with Anthem. And so it was merely a matter of reconciling those on the portal as is presented to the client ordering. Which like I said, I think in the last few weeks, we've got that mostly cleaned up. Our teams are kind of watching it, and we're working with Anthem to make sure that it -- goes more smoothly in the future. And I think the only other question, one I didn't answer was the percentage of Anthem business affected. And I don't think we have that off the top of our head. Okay.

Just in general, it did double basically the denials of the tests. And so say, we were about 25% denial rate, it was something like 50% for the third quarter. And we do expect, as Sean said, to have that reverse back to that prior level once we work through these issues in coding and such within the system.

Okay. That's really helpful. And just to be clear, because volumes were about what we expected, but ASPs were less than expected, I just want to be clear. Were the volumes done? You just didn't get reimbursed. Or were volumes actually impacted here?

Yes, so volume -- as we kind of have said in the past and indeed expected, Q3 tends to be a light quarter compared relatively to the other quarters. And this Q3 was no exception. Now the biggest drivers are the summer break; and then, wherever NSGC tends to land, at least at this point in the company's lifespan, a good chunk of our customers go to NSGC and it typically takes a good 3 to 4, maybe even 5 selling days out. Those both coincided with this quarter. So we were expecting about the volume we've got. And then, the revenue -- on the revenue side again, this is now moving to things like Anthem. But really, this is about the execution of our in-house billing. As Shelly mentioned, we've now completed bringing it in. We are performing on the core billing metrics. You could think of all manner of looking at speed of billing and collectibility. We're performing as good as our prior vendor had, and that was after 3 years of working with us. So we're pleased with where we stand today on tangible metrics we can look at, and we would expect -- again, we expect that to pick up as time goes on.

The only thing I would add is that, we do not try to predict and give guidance on a quarterly basis. And I would say that our second quarter was a little higher than we would have anticipated. And we probably pulled a little bit of that revenue in to the second quarter as opposed to having it fall in the third quarter. And so it's very tough when you try to judge these things on a quarterly basis. We look at it more as a trend. And so we believe it's all based on improving the collections over time, and that we are on track to do so.

Okay. So last one, and recognizing what you just said, Shelly, and recognizing you don't guide by quarter, but implicitly you did because of where we are in the year. At the midpoint of guidance, it appears that your expectation for core ASPs in the fourth quarter, basically that you're assuming a jump back into the low to mid-600s. Is that right? And if so, what drives that increase? Because mathematically, if we just look at your volume and your revenue guidance, that's where we come out.

Yes, I think in general, that's the right neighborhood, Doug. That's -- I mean, like you said, we are now -- implicitly in our annual guidance, you now have our Q4 guidance. And it sounds -- that sounds about right to us.

Okay. And that's just based on typical seasonality in timing and collections.

That's right. Well, the volume is mostly seasonality. And then, well, I guess, the improvements on our in-house billing operations. Lee is...

There's also just a little bit of catching up, Doug. So some of what resulted from the implementation of in-house billing resulted in some timing issues with revenue that might have come in, in Q3 getting pushed into early Q4. So that -- this is a function of the fact that we're still largely cash. So I think that's the other factor that will get us into the range that we have guided to.

Your next question comes from Puneet Souda from Leerink Partners.

Sean, the ECS offering, when can you introduce that into market? And what do you think? What would it take to get back your market share there?

Right. So we expect -- we're obviously working at Invitae's speed on this. And we would expect in the first half of next year to have an Invitae version, which will be broader, faster, less expensive, all things clients have come to expect from us. The Good Start commercial team has great relationships with these clients. They've built a really -- again, they've built a real quality business, a tight business with these IVF clinics. As Shelly mentioned, we're expecting for a period of time not to have that ECS volume. The rest of the volume that we've acquired from the Good Start business, it's unclear at this point what exactly is going to happen. I think optimistically, we're holding it and keeping it. It will be challenging until we have a replacement for that. I would kind of -- I think, macro level, this is kind of the stuff you expect. And I think, of course, the -- our path forward with the company in decision-making in it, would kind of have this stuff prepriced as it were. But again, we expect to get an Invitae version of this early in the year, and we'll continue on the swift commercial execution that we've got off to a good start on.

Okay. And then, on the pre-auth headwind that you're going to have here, I mean, just help me understand. Obviously, you have Anthem and United now, and a number of major payers. So help us understand if that continues to expand here into next year, how should we be thinking about the overall macro impact on genetic testing here?

Yes, I think kind of, again, the macro impact is as we have kind of postulated and, I think, stated over the last 18 months, 2 years, the insurers are kind of coming around to genetic testing. I mean, it's obviously one of their fastest-growing line items, which has caught their attention; historical prices, second, kind of sharpening their focus and attention. And now you see them putting in place similar mechanisms that they have. It's very common these kind of -- automated pre-auths with many pharmaceutical in high-growth or high-value line items for the insurers. So on a macro level, if you take a step back, the biggest implication is that as we have suspected and suggested, life as an out-of-network provider will get harder, and life as an in-network, trusted provider of goods and services will get a little bit easier. And we'll continue working with them as much as we can to make sure that Invitae is the preferred partner therein.

The only thing I would add is that we are able -- since the billing is in-house, we are able to improve upon our automation of those prior auths, not only through the existing systems, like an AIM or a Beacon, but also improving our prior auth across the rest of the industry and the third-party payers. And so that helps us to ensure that more of the tests will actually be collected on the back end. It's an improvement on the front end that yields good results on the back end. So we like using technology to enable us to continue to automate that prior auth process, not only through them when they're forced upon us, but also on our own.

Okay. And then, on the quality control, I just want to understand, what are the steps you're taking? And where do you stand right now? And what's -- help us understand and help us get comfort around how these QC issues can happen in the future. I mean, this is a fast-evolving landscape, and I know you're adding tests to the menu all the time. So what are new practices in place that can potentially prevent this in future again?

I'd say that to summarize it, we've basically tripled down on the documentation and any kind of implementation of revisions in the kind of target areas of the genome. That's the best way to put it, kind of without getting too much in the nitty-gritty about the bait sets and the [pro borders] and whatnot. Essentially, we -- kind of acknowledging that we're now getting to a scale where these issues are -- kind of have these kind of ratifications. We haven't always been there. We are there now. We've implemented additional quality control checks, implemented and are implementing additional documentation and change control management, that certainly for this issue and just about every issue kind of in relation to this kind of thing, will prevent that, will prevent that from happening again.

Your next question comes from Tycho Peterson from JPMorgan.

A couple follow-ups. I hate to hit on the pre-auth thing again. I just want to be sure I'm clear. On the fourth quarter, you were saying kind of no residual impact, is that right? And then, I think if I interpret your comment right, Sean, you're assuming that other payers kind of follow the lead of Anthem and United next year. So should we think about more of an impact in 2018?

Yes. I think 2018, the impact is probably neutral, right? I think each -- payer by payer, there's going to be some rollout and follow-up. Again, we believe the Anthem AIM rollout, when you think of all the issues there, we believe we worked through most of them. So we don't expect a material Q4 impact on revenue from Anthem. United is going well enough that we don't -- we're not calling out anything specific we foresee in United. The others, I think, our plans and you can kind of see the third-party vendors that's doing -- pulling these kind of things together, getting ready to implement. I believe there probably has been an industry-wide, kind of lesson learned between both the Anthem and the United rollout. And so no, I don't think there's, at this point in time, we're certainly not going to call out any difference in our view of revenue based on these. We really do think -- I think as Shelly mentioned, the key second point there is once they're up and running, billing velocity and mechanics go much, much better. So it actually -- once they're up and running, it actually is better for us. So I think in kind of net-net on a 2018 perspective, this is a nonstory, at least, as we sit today.

Okay. And then on PAMA, appreciate you guys sending out letter to us as well. Maybe it was 2 weeks ago you sent it in, I guess. Any feedback from CMS? And what gives you confidence that, particularly those 2 codes, 81432 and 81435, do get overturned?

Yes, so we have been in constant communication with CMS, amongst others. We kind of gotten the -- what you would expect in terms of exchange, acknowledgment of receipt and interest in working towards the correct outcome. We are generally under the impression that the good folk at CMS are indeed scrambling to figure this out. That's our general understanding. With that said, we haven't gotten any clear indication of final resolution, so we're kind of in a wait-and-see mode at this point, which is why I point out, worst case here, what the expectation could be. But again, it seems so obvious to most involved that we're optimistic there will be some kind of resolution to iron this out.

Okay. And then, you guys mentioned you began cross-selling Good Start and CombiMatrix in September. Any early feedback from some of the initial rollout?

Sure. Tycho, we did roll that out in October. And what I would say is that, it really -- the signal, based on the feedback that we've gotten from customers, that the combination of the testing (inaudible), moving forward, is going to play really well for us in terms of being able to be the comprehensive provider to these clinicians. And I think that really sets us up well for 2018. And plans are moving ahead in terms of sales force integration and training. So I would say that the early feedback is a really good signal for what's to come.

Okay. And then, just last one on the Lynch syndrome retesting. Now that that's complete, looking back, any implications in terms of ordering physician relationships? Any kind of ancillary damage control you need to think about now that you've kind of resolved the issue?

No. Yes, that's a good question, glad you asked. In fact, honestly, coming into Q4, I actually feel our brand has actually been, not only preserved, but indeed enhanced. This is some pretty direct and frank feedback from clients about what their expectations are in the industry. Kind of that these things tend to happen, and that the way we handled it, they felt, was exemplary. And again, like I mentioned, I'm incredibly proud of the team's response. I'm kind of proud of our latent technical capabilities to get on top of these things and quickly. The expertise of our medical staff, I believe is unparalleled in the space. And as you work through this kind of issue with our clients one by one by one, I walk away with a perspective that there's a -- I mean, sure, there are some cautionary points and make sure you guys don't mess this up, and that kind of thing. But on the whole, I've walked away from this experience feeling that we certainly did the right thing by the patients. And indeed, and as one might expect, our clinician clients have also gone positive in their regard for us. So I think while, of course, we would never indicate this is a positive development. This was a manufacturing mishap. This is our first one that was really kind of publicly on display. With that said, I think the team, in our approach, just demonstrated exactly what Invitae's made of. And I think it just really harbors good things for the future.

Your next question comes from Raymond Myers from Benchmark.

I want to first ask you, Sean, about the effect of prior authorization on physician prescribing behavior. And has it had any impact on shifting which companies' tests they are completing the order for?

Right. So I think I would break this down into 2 time frames: One, in the time frame we're talking about, which I would just call it general confusion in physicians just trying to figure out how to order the test that they want to order. And in that time frame, I don't think it shifts from one provider to another. There's just a slowdown in the whole thing. There's a lot more phone calls. There's a lot more client service impact. There's a lot more field reps going into the accounts and handholding, trying to figure out what's going on. And ultimately, again, it's not like whoever they were going to order from, it's not as if they shifted it. It is just there was a ton of confusion around it. And then, whoever got the order, couldn't get it paid for because that it came through -- it was essentially it wasn't a clean bill from the beginning. And so let's call that in the Q3 time frame. I don't think it shifts a lot between competing companies. I think it just kind of bogged everything down and, of course, as we indicated, was likely a revenue hit to all. And I mentioned, to the extent we can quantify it, it was around 400k for us. Now going forward, this is where I think -- going forward, as it gets all cleaned up -- and in fact, if you kind of use the Beacon example as perhaps a little more illustrative of how this will work in the future, only in-network companies are listed. And they are listed by kind of the volumes that are ordered, the value perceived by the insurance, the commercial insurance organization. And so I think over time, this serves, indeed, as we have been suggesting for a while, this will serve as a tool for the insurers to indeed divert volume to the labs that are the most comprehensive, highest quality and best value for the dollar. And we think that over time, this benefits us above most.

Very good. How many salespeople does Invitae have today?

Sure. So we -- our sales force is about 50 folks today, including our inside sales team. So come January 1, we'll have a fully integrated sales force of about 100 individuals. So that is what the team will look like for 2018. We're beginning to do our sales force training to ensure that all of them are up to speed in terms of how to talk clinically about our cancer offering, our reproductive health offering. And then we'll continue to have a team focused on children's hospitals. So they're all very excited about the new content. That's something that they see a kind of enthusiasm about from their clinicians as they started talking about the future of Invitae. So I would say that the enthusiasm on the part of the sales force is really high for 2018.

Great. I understand that you've made a lot of improvements and continue to, in your lab, to keep driving down your cost per test. We saw more progress in third quarter. How much more progress do you expect throughout, say, the course of 2018 in driving down the cost per test?

Right. So this is one that I can, I think as we've begun indicating and I think would kind of reiterate today, we continue to invest in our technology leadership, in particular, the ability to extend our cost advantage. With that said, as that proceeds, we're also bringing online other tests and other operational requirements that come into play. And maybe Lee can kind of give a -- kind of a view of what 2018 and beyond will look like in regard to COGS in the operations of the company.

So there are a number of opportunities ahead of us to reduce the average cost per test. We are, for example, just now implementing Novaseek across our entire test menu. The real thing that we are debating right now on, honestly, is -- and I think you started to see it a little bit in this quarter. We have now gotten to the place where, as we approach our 50% gross margin target, the choice will be to -- whether to continue to drive down the average cost per test report, or to make our answers more comprehensive and add content in ways that will be useful to clinicians without increasing the average cost per test and while maintaining our target margin. And so in each case, it'll be kind of a tactical decision about what is best for the patient and what enhances our lead over the competition. So you can expect to see both with just volume as well as continued focus on things like Novaseek and other technology enhancements to see COGS go down. But I think that it'll, frankly, start to slow as we look at opportunities to add more content and provide more comprehensive and more useful information to clinicians and their patients.

At this time, I'd like to turn the call back over to Sean George for closing remarks.

Great. Again, thank you, everybody for joining us for the Q3 call, and we look forward to catching up with everybody in future conferences. I'm excited to get to our annual call in February of next year. Thanks again.

That concludes today's conference call. Thank you for participating. You may now disconnect.