We appreciate the efforts CMS has undertaken as they work to finalize the coverage policy for NGS testing for advanced cancer broadly and for FoundationOne CDx specifically. Yesterday, CMS indicated that they expect to finalize the coverage policy by March 16, allowing for a full 60-day administrative window, following the comment period that ended on January 17. We remain encouraged that the final policy will provide eligible Medicare beneficiaries with covered access to well-validated diagnostic tests, like FoundationOne CDx. On today's call, I'll review several key highlights from 2017, provide insights into our 2018 strategic priorities, and then turn the call over to Jason, who will review our financial results and our 2018 guidance.

To begin, clinical adoption remained strong throughout 2017. We reported record volume of more than 67,000 cases. That's a 54% increase year over year. Robust adoption of our CGP approach was driven in part by increasing number of actionable, targeted, and immunotherapy biomarkers on our assays, including tumor mutational burden, or TMB, and microsatellite instability, or MSI.

Our full suite of CGP tests are providing oncologists with critical molecular information to help inform patient-treatment plans, including targeted therapies, immunotherapies, and potential clinical trials. Our biopharma business continued to grow and diversify as well, creating value across the entire drug-development life cycle from target discovery to commercialization. This work with biopharma not only contributes meaningful revenue, but it also inherently benefits our clinical business by expanding the relevant choices for therapeutic treatment options for physicians over time. We signed new and expanded partnerships, including with BMS and most recently with Pfizer, further establishing Foundation Medicine as the molecular-information partner of choice, particularly in the area of immuno-oncology.

Importantly, and as a testament to our diversified business, in the fourth quarter, our biopharma revenue grew by 75% over the same period a year ago and was driven by our non-Roche pharma partnerships. Additionally, these partners also continued fueling our R&D pipeline in 2017. As an example, we presented data demonstrating that our novel assay measuring TMB from blood could be used to predict the response to immunotherapy. It is being used in Roche/Genetech's Phase III Blood First Assay Screening Trial, or BFAST, as a potential companion diagnostic for multiple targeted therapeutics in advanced lung cancer.

And most recently, BMS announced through its Checkmate-227 trial utilizing FoundationOne that TMB is an important, independent, predictive biomarker to identify a population of first-line, non-small-cell lung cancer patients who may benefit from combination therapy. This is a meaningful advancement, as it's the first time that we're able to demonstrate prospectively that TMB has clinical utility as a biomarker for immunotherapy. We also continue to create value across the oncology landscape from our molecular-information database, and that includes growing our clinical genomic database offering with Flatiron. This combined database has more than 33,000 match profiles, which is the largest of its kind in oncology.

We look forward to growing this partnership and providing valuable R&D insight to even more biopharma partners. Operationally, we commenced sample processing in our lab in Germany, and in partnership with Roche, we've now launched commercial activities in 27 countries, spanning parts of Europe, South America, the Middle East, and Asia. Our expanding presence is important to our clinical business over time, and our biopharma partners highly value our expanding footprint for their global clinical trials. But most notably, we've capped off a strong year with FDA approval and preliminary NCD for FoundationOne CDx.

FoundationOne CDx is now the first and only comprehensive genomic profiling test for all solid tumors to receive FDA approval. The preliminary NCD provides coverage for FDA-approved companion diagnostics for the management of cancer patients with solid tumors that are metastatic, including stage IV and recurring cancer. This is expected to materially expand coverage for eligible Medicare beneficiaries nationwide by initially covering five tumor types that include lung, breast, ovarian, colorectal cancers and melanoma. Importantly, the NCD is designed so that coverage can readily expand as companion diagnostic claims for new tumor types are added over time.

There are more than 600 unique oncology drugs in late-stage development, many of which are expected to require a companion diagnostic for approval. The five covered tumor types account for approximately 50% of our historical FoundationOne volume. As a reminder, Medicare and Medicare Advantage patients represent approximately 40% of our testing. The preliminary NCD calls for Coverage with Evidence Development, or what's called CED, for all other late-stage solid tumor types.

And that's when the patient enrolls -- is enrolled in a clinical study or registry. This could represent an additional path for coverage for FoundationOne CDx for Medicare beneficiaries with all late-stage advanced solid tumors. On the heels of our many accomplishments in 2017, Foundation Medicine is poised to continue growing the business while changing the landscape of cancer care as we know it today. To that end, I'd like to outline our strategic priorities for 2018.

First, we plan to commercially launch FoundationOne CDx once the NCD is finalized. In tandem with the launch, our commercial team will continue driving awareness of our full suite of products for solid tumors, hematology cancers, and a liquid biopsy when tissue is not available. All of which enable oncologists to select the right test for the right patient with confidence. In addition, our global expansion efforts will continue to broaden in partnership with Roche.

Second, we will operationalize the NCD for FoundationOne CDx to secure Medicare reimbursement once the NCD is final. Our unique PLA code was recently issued and will become effective on April 1, and we're awaiting official guidance and designation as a new ADLT. In addition to securing payment for Medicare cases under the NCD, we will work with private commercial payers on both their Medicare Advantage and private-covered lives. An important initiative supporting our efforts for payer coverage is the inclusion of broad profiling in clinical-care guidelines.

Most recently, NCCN expanded guidelines to include molecular testing for ovarian and prostate cancer as well as melanoma, and we expect more to follow. The evidence for taking a CGP approach is highly compelling today and will only continue to grow as more targeted and immunotherapies are approved. Our third objective is to cultivate new and expanded biopharma opportunities with those working in oncology drug development. Demand across all of our solutions remains strong, particularly companion diagnostic development given our FDA approval, which has further strengthened Foundation Medicine as the CDx partner of choice.

Most recently, we signed a broad collaboration agreement with Pfizer for CDx development and data access, and we expect to sign new and expanded agreements throughout 2018 as we continue to serve as a powerful innovation engine for our partners. This brings us to our fourth priority, which is to continue to leverage our data to shape the targeted and immunotherapy landscape with new and emerging biomarkers. It is through our proprietary data and algorithms, as well as working closely with multiple biopharma partners, that we drive innovation and remain on the leading edge of biomarker discovery, and ultimately, expand the relevant treatment choices for physicians and the patients for whom they care. This is the power of our synergistic and unique business model.

2018 promises to be an exciting year. We believe that we're just getting started in terms of the impact that we can have on the lives of advanced cancer patients and their loved ones while helping to deliver on the promise of personalized cancer care. With that, I'll turn the call over to Jason to review our 2017 results and provide financial guidance for 2018.

Jason Ryan -- Chief Financial Officer

Thanks, Troy. As a reminder to everyone, we announced preliminary results in early January, so I'll briefly review the fourth quarter and full year 2017 and enter into the outlook for 2018. For the fourth quarter, total revenue was $48.9 million, a 70% increase over the $28.8 million recorded in the same period last year. As a reminder, starting in Q2 of 2017, we began reporting revenue in two components: molecular information services and pharma R&D services.

Molecular information services is revenue derived from commercially available platforms, such as sample profiling and SmartTrials, and includes revenue from both clinical and biopharma customers. Pharma R&D services is revenue derived from our work that is funded by biopharma to develop new and expanded offerings. In the fourth quarter, revenue from Molecular information services was $37.4 million, an increase of 83% year on year. This includes $15.5 million in clinical revenue versus $9.8 million in the fourth quarter of 2016.

The increase was driven by overall clinical adoption, as we reported 20,044 cases -- clinical cases in the period, a 57% year-over-year increase. The average revenue across each clinical CGP test recognized in Q4, excluding international volume, was approximately $2,600, consistent with the third quarter. Molecular information services in the fourth quarter also includes $21.9 million in pharma revenue versus $10.6 million in the same period last year. This substantial year-over-year growth was driven by a particularly strong fourth-quarter sample profiling, as we reported approximately 6,200 tests across our biopharma customers.

Our pharma R&D services in the fourth quarter was $11.5 million, as compared to $8.4 million in the same period last year. All together, total Q4 revenue from our biopharma customers, including both molecular information and pharma R&D services, was $33.4 million, as compared to $19 million in the fourth quarter of 2016. Our OPEX in the fourth quarter was $59.9 million, an increase from $56 million in the third quarter, primarily driven by nonrecurring G&A expenses. For the full year of 2017, total revenue was $152.9 million, a 31% increase from the $116.9 million reported in 2016.

Revenue from molecular information services for the year was $117.2 million, which includes $53.2 million in clinical revenues and $64 million in pharma revenue. In pharma R&D services, revenue was $35.7 million in 2017. Turning now to our 2018 full-year outlook. We expect total revenue to be in the range of $200 million to $220 million, or a 37% year-over-year increase at the midpoint.

Revenue from molecular information services is expected to be in the range of $160 million to $175 million, or a 43% increase at the midpoint. Within that, we have a new category. We expect clinical revenue will increase approximately 70% to 90% year over year, driven in part by anticipated Medicare payments under a final NCD and by increased clinical testing volumes, which I'll turn back to in a moment. Also within that revenue category, we expect pharma revenue will increase approximately 10% to 20% year on year, driven by sample profiling, data insights, and SmartTrials.

Revenue from pharma R&D services in 2018 is expected to be in the $40 million to $45 million range, or a 19% increase at the midpoint, driven by our CDx partnerships as well as important funded R&D initiatives. Related to the timing of total pharma revenue in 2018, we anticipate a sequential decline in Q1, due solely to the timing of various work streams. This is consistent with our experience last year, and we're confident that pharma revenue will continue to ramp over the course of 2018. We expect to report in the range of 90,000 to 100,000 clinical tests, or a 41% year-over-year increase at the midpoint.

In addition, assuming we remain on track with the finalization of the NCD and our new ADLT designation, we expect to transition the majority of patients' cases from FoundationOne over to FoundationOne CDx over the course of Q2 through Q4 of this year. We anticipate gross margin from molecular information services to be roughly 40% for the year. Keep in mind, that includes the cost of unpaid clinical tests, which does weigh down the average. From a product gross margin standpoint, we anticipate a gross margin of roughly 70% on paid FoundationOne and FoundationOne CDx tests.

We expect OPEX to be in the range of $250 million to $260 million, an increase of 10% to 15% year on year, as we continue to invest in systems infrastructure, product development, and commercial execution. These 2018 investments are specifically designed to support our top-line revenue goals and also to drive additional leverage, volume, and revenue growth in 2019 and beyond. Our CAPEX is expected to increase year over year in 2018, driven by expansion of our facilities in both Cambridge and RTP to accommodate additional regulated capacity as well as for business-continuity purposes. We began 2018 with approximately $70 million in cash and cash-equivalents and $140 million remaining under the Roche credit facility.

This $210 million puts us in a good position from a capital standpoint. On a final note, as of January 1, 2018, we transitioned to the new rev-rec standard, ASC 606. Under the new standard, among other changes, we've begun recording clinical revenue on an accrual basis as opposed to the cash basis we had been using previously. The impact of this change is fully incorporated into our 2018 revenue guidance.

There will be a one-time transition adjustment in Q1 to reflect the estimated future payments on claims that were outstanding as of 12/31/17. This adjustment will hit retained earnings as opposed to revenue and, therefore, is not part of our 2018 revenue guidance. During 2018, we'll be disclosing on a quarterly basis both the accrual and cash-based clinical revenue. Additional important information on the new rev-rec standard is included in our 10-K.

In closing, Foundation Medicine had a very strong 2017, both in terms of financial growth and also in terms of the significant milestone with FDA approval and the preliminary NCD for FoundationOne CDx. We believe that we're well-positioned for continued growth and that we can both meaningfully and uniquely expand patient access to personalized cancer care in 2018 and beyond. And with that, I'll turn the call over to the operator. Operator?

Questions and Answers:

Operator

[Operator's instructions] And our first question today comes from Amanda Murphy from William Blair. Please go ahead with your question.

Amanda Murphy -- William Blair and Company -- Analyst

Hi, thanks. So actually, Jason, just a follow-up to your commentary on guidance. So I guess, how are you thinking about the NCD rolling online, given that you obviously don't have the final yet? And I'm assuming that there'll be some logistical dynamics around payment or whatnot. So how to think about that? And then any sense that you can give in terms of what your -- I'm sure, and I think it was like a quantified number, but just how much of the guidance -- how much of the revenue guidance is related to NCD, if that makes sense.

I'm not asking about profit. Hopefully you get the sense of what I mean.

Jason Ryan -- Chief Financial Officer

Absolutely, yes. No, it makes sense. Thanks for the question, Amanda. So in terms of the NCD coming online, remember there are a few pieces that would come together, right? So one is finalizing the NCD.

That's the first part. The second is finalizing the new code for FoundationOne CDx, which we have done. So we have a code for F1CDx. And the final part, as Troy alluded to in prepared remarks and I did as well, is ensuring that we're designated as a new ADLT.

We certainly believe we are, and our expectation is we'll be a new ADLT. With those in place, and in particular, with the new rev rec, potentially even without it, we would begin accruing revenue for FoundationOne CDx cases. Obviously, important in the equation for this year is the transition from F1 to F1CDx over the course of the year. So once we have these three pieces, the final NCD, the code, we've checked off, and the new ADLT designation, then, at that point, as we're reporting out FoundationOne CDx cases, we begin reporting revenue.

Obviously, because of the transition from F1 over to F1CDx, the more material part of the NCD revenue will be in the back half of the year because that transition won't all happen at once. So while we're not breaking out the sort of specifics of NCD revenue, that's how we're thinking about it in the model

Amanda Murphy -- William Blair and Company -- Analyst

And I guess, said another way, are you -- obviously, you're getting paid for lung now, so I guess a two-part question for that. What happens to the LCD component as you transition to the NCD? And then how much of your guidance assumes payment for Medicare above and beyond lung? Are you assuming -- is there a way you can sort of frame out how much of it is beyond lung?

Jason Ryan -- Chief Financial Officer

Yes, I think one of the advantages of taking this sort of dual-path strategy that we have of approaching both the local MAC as well as CMS through a National Coverage Determination is we have, I wouldn't say optionality, but we have flexibility here. So obviously, moving to the broader coverage under the NCD is important. We are all going to need to wait to see what the final legislation is and the rules are around LCDs and whether they exist or not or over what period of time. That LCD piece for us is less material for the year.

The more important piece is moving over to the NCD. So we do have in our guidance, we have included some LCD revenue, but it is not a material amount over the entire course of the year, given that we're transitioning over to F1CDx. So we've taken all of that into consideration in this 70% to 90% clinical revenue growth.

Amanda Murphy -- William Blair and Company -- Analyst

OK, got it. And then just, I guess, the last one. So obviously, you talked about TMB and the data that Bristol presented for that marker. It feels like there's a lot of markers in the works, either that have been approved, if you think about KEYTRUDA and MSI, or that are in the pipeline that are kind of cross indication, if you will, or without a specific indication.

So can you talk about the process by which you would amend the PMA to be able to incorporate the types of markers? And is it realistic to say that you might be -- I don't know over what time frame, but your Medicare coverage really expand under the way the NCD is structured just given that Medicare is covering whatever label you have under the PMA?

Troy Cox -- Chief Executive Officer

Sure. Hi, Amanda, it's Troy, I'll take that one. Of course, we're excited about the design of the Medicare policy, and it travels together with the -- with how the data is, quite frankly, really exploding in terms of more and more. As you well know, we have MSI today.

Certainly, it's reasonable to assume that TRK fusions and TMB downstream will be biomarkers with approvals, it's certainly reasonable to assume, agnostic to tumor type. But the way that we will -- the way that the model will work is that we can work with the FDA to get CDx approval within a specific tumor type. And once that is -- it's essentially just like amending a drug label with a new indication, we would amend our label of the first and only approved comprehensive assay for all solid tumors. We would amend that with the new CDx.

And then that would, in turn, provide for increased payments through reimbursement in an unrestricted way in the NCD portion versus the CED portion of that Medicare policy.

Amanda Murphy -- William Blair and Company -- Analyst

OK. And I mean, if these markers are pan-cancer, wouldn't that then mean, implicitly, you would sort of have coverage across all cancers? [Inaudible]

Troy Cox -- Chief Executive Officer

The preliminary NCD -- yes, understood. The way the preliminary NCD is written today, indeed, that would be the case. That's our understanding.

Amanda Murphy -- William Blair and Company -- Analyst

OK. Thanks very much, guys.

Troy Cox -- Chief Executive Officer

Thanks, Amanda.

Operator

Our next question comes from Doug Schenkel from Cowen. Please go ahead with your question.

Doug Schenkel -- Cowen and Company -- Managing Director

Great. Good afternoon, guys. Just to start with, I guess, a couple more guidance, clarification questions. I guess the first one's really on guidance philosophy.

I just want to be clear, your guidance incorporates a clinical volume assumption that is based on the belief that the NGS NCD goes through as proposed. Is that correct?

Jason Ryan -- Chief Financial Officer

That's correct, Doug. So our base case here is that the NCD does go through. And we're guiding both basically on volume and revenue, assuming that that's the case.

Doug Schenkel -- Cowen and Company -- Managing Director

OK. And then, in terms of pricing associated with the NCD, I know Amanda touched on a little bit of this. But in our models, just to be clear, should we assume CMS ASPs on tests to reimburse for around $3,000? As you know, you have some flexibility on ADLTs for, I believe, the next -- might be three quarters, where you could book up to list. But I believe you've indicated that we should be modeling something around $3,000 as we've talked about the 50 gene panels and what we believe Palmetto's going to pay for FoundationOne.

So is that $3,000 number the number that you've embedded into your numbers, your expectations?

Jason Ryan -- Chief Financial Officer

Yes. I think that it -- while I won't mention exactly what's embedded in our expectations, I will tell you that $3,000's a good number. We shared that a couple of months ago, as you know. We said at or above $3,000.

That's what we mentioned. And just so everyone -- and I know you understand. This is for others listening in, right? Under PAMA, CMS will pay us a rate that's based on our average allowables from the third-party payers. And during this few quarters, you mentioned, we need to get experience on billing those third-party payers.

During that experience sort of period, we can bill CMS, and we'll get paid at a rate we set ourselves. The reason companies may not and we would not bill at our list price is that there's always a clawback at the end as well as that's just not the price we anticipate needing for the business. And so, yes, we'll be looking at or above $3,000. Sorry for the long answer, but I want to make sure that everyone has the background.

Doug Schenkel -- Cowen and Company -- Managing Director

That's super helpful and makes sense. Maybe just a couple, I guess what are really Roche-specific questions. I think it was earlier this week that it was disclosed that you had amended the commercial agreement with Roche. Looking through it, it appears that the big changes were largely the inclusion of FoundationOneCDx and maybe I'm reading into it too much, but it seemed like there were some at least adjusted or new language that seemed to be essentially there to make sure potential CDx partners know that your distribution from Roche isn't gonna favor Roche pharma over others based on the fact that the Roche reps working on FoundationOne aren't gonna overlap as drug reps.

Was that the intent or are those, cutting through all of it, were those really the major changes to the agreement?

Jason Ryan -- Chief Financial Officer

The primary change, Doug, you hit on, which is really including FoundationOneCDx under the portfolio of services that they can offer, important, I think, all the way around, both with Roche and with us is that we continue to work with our pharma partners, Roche and not Roche, really globally and around the world, I think that's an important, sort of an important guiding principle to how we work with Roche on most things. And that is, so if you saw some of that in that amendment, it would follow that guiding principle.

Doug Schenkel -- Cowen and Company -- Managing Director

Got it. OK. And last one, really I guess, to some extent, building off of Amanda's question on how you would evolve FoundationOne CDx as you potentially add new biomarkers or want to add new biomarkers. My -- internally at Cowen, my biopharmaceutical partners tell me that Roche pharma is talking more and more about TEF, and I believe they've positioned that as a somewhat proprietary-developed biomarker.

In instances like that, whether it's TEF-specific or something else, if it comes from Roche, given your relationship with them, would that be something that would be easy to add onto your menu, if you so chose to do so?

Troy Cox -- Chief Executive Officer

Yes, we'll have Vince add some color. But I just want to make sure that everyone's grounded on what's in our test today, important biomarkers that are driving really important decisions today across our existing portfolio of products. And as you recall, when Merck's pembro got approval for agnostic to tumor type for MSI, we already had MSI in our test. And we, on Page 1 of our report, we were guiding oncologists for those patients with an MSI positive to consider that as an option.

We already include TMB within our test as well as we're able to capture TRK fusions as well. And so that's -- the value of our comprehensive test is that, indeed, we have a lot of the biomarkers that are common today that there's solutions identified for, that we're already providing guidance and a molecular fingerprint, if you will, for the patient and for the oncologists. I'll turn it over to Vince, if he wants to add some more color on specific biomarkers.

Vincent Miller -- Chief Medical Officer

Doug, apologize for my line and -- if it's echoey. But I think you're asking about the T effector signal or signature in I-O predictive [Inaudible]. That being said, of course, we're partnered with most companies with significant portfolios in this area, and each has a slightly different or substantially different strategy. That's sort of the beauty of where we sit at this sort of epicenter of things.

And similarly, if another partner, BMS or someone else, had another predictive biomarker of interest, we would be greatly positioned to bring that forward, first as an R&D assay and then as a clinical product, if it were validated.

Doug Schenkel -- Cowen and Company -- Managing Director

Great. OK. I will -- and I'll leave it there. Very helpful color.

Thanks, guys.

Troy Cox -- Chief Executive Officer

Great. Thanks, Doug.

Operator

Our next question comes from Patrick Donnelly from Goldman Sachs. Please go ahead with your question.

Patrick Donnelly -- Goldman Sachs -- Vice President

Great. Thanks, guys. Maybe just one on gross margins. I mean, I know, obviously, gross margins will be mostly driven by cases covered going forward.

And covered tests seem to be stable around 70%. So maybe just expectations there? And then, are we still thinking costs for tests around $800 for 2018? And then can you just talk to the opportunity to lower that -- the half that's variable, the cost there, what are the biggest levers?

Jason Ryan -- Chief Financial Officer

Certainly, yes. This is Jason. So we provided some gross margin on the molecular-information line just to make sure that the income statement worked from a model standpoint. What we are focused on, as you mentioned, is gross margin for those paid cases, and we do think that gross margin will be roughly 70%, give or take, on FoundationOne and FoundationOne CDx.

That calculation is at a modest all-in price across those paid cases. With respect to COGS, where we've recently been in a range of $800 to $900 fully loaded for FoundationOne, we do believe there's an opportunity for that to come down over time. We don't see that cost getting into some range that could be $200 to $300. We do think it will come down with some increased -- with some additional efficiencies and [Inaudible] technologies.

And also, there are still economies of scale that we think, as we continue to scale, we'll see. So we do believe costs can come down over time. And really, what we've said historically and still continue to believe today, is that anywhere in a range of $3,000 for a FoundationOne test puts us at or above the 70% gross margin. And then we've got some, we believe, longer-term favorable dynamics on OPEX when you think of operating margin longer term because of the sort of impact and leverage that our commercial team has selling a comprehensive universal suite of products.

So I guess, I'll leave it there. Some COGS flexibility in the longer term, and a very strong margin profile anywhere around the $3,000 mark.

Patrick Donnelly -- Goldman Sachs -- Vice President

OK. And then maybe just on the biopharma partnership side, I mean, this seems to be getting more and more a compelling part of the story. But can you just discuss where you think the number of partnerships can go over next year or two? I'm just trying to size the opportunity in my mind. And then are there one or two partnerships -- I know you mentioned a couple on the call, that you'd call as particularly meaningful the next year?

Troy Cox -- Chief Executive Officer

It's Troy. Indeed, we're over 30 biopharma partners today. And while we're not giving specific guidance on how many specific partners -- and I don't think that's the most important metric as well, although we do believe that we'll grow in total number, that's a reasonable assumption to make. We've also made substantial strides with more kind of breadth and depth within our existing partnerships.

And so quite frankly, we've migrated to approach over the course -- started this really in late '16 through '17 -- is we've taken a bit of a cafeteria style with our biopharma partners, and they would pick one or two items. And so now we're really -- have seen success, rewarding for both our biopharma partners and us, building more substantial partnerships, as we've done with BMS and Pfizer, is the model that we'll pursue. And I think with confidence, I can -- I'm confident that we will continue that success. The really important thing to understand here is that as a CDx partner, we offer multiple sources of value.

But it just doesn't get any more attractive. We're the partner of choice for CDx because we can do it faster, better, more cost-effective. And most importantly is, we have the leadership position and market to be able to pull through and make sure the doctors have access to the right tests and, therefore, helping our biopharma partners identify patients for their therapeutic therapies. And so I feel really good about the potential of continuing that approach and that strategy with our biopharma partners.

Patrick Donnelly -- Goldman Sachs -- Vice President

OK. Let me just sneak in one last one. I mean, I know International feels like a bit of a longer-term [Inaudible]. A lot to still do in the U.S.

But just thinking about how you can leverage Roche's sale force OUS, I mean, should we be thinking about that as pretty minimal impact in 2018, and then more of a kind of '19, '20 story?

Jason Ryan -- Chief Financial Officer

Yes, I'll take a first stab, and then Troy can add. Patrick, it's exactly the right way to think about this. It's strategically very important for us, not only in the clinical side, but also on this pharma side, as Doug alluded to in his questions. It's very important pharma ex U.S.

We do think we're going to start seeing some more meaningful volumes in '18. The story is really is about '19 and beyond, so we have all these countries that now have some infrastructure getting up and running. As you know, every one of these countries is different when it comes to KOLs and regulatory etc., and commercial forces. So I think 2018, look at that as the year where we're starting to really get momentum in those countries that we have set up to lead to really greater, if you will, ubiquity over time outside of the United States.

Troy Cox -- Chief Executive Officer

Yes. And the other thing I would add is that, as we've referenced already, we're in 27 countries. So certainly, we expect to -- there's strong support from both Roche and Foundation Medicine to continue that effort of expansion in 2018. When I reflect on this, if we just pretended for a second that we didn't have the Roche partnership, this company would have a hard time -- Foundation Medicine would have a hard time expanding globally.

And so it's a real great -- one of the many rewards from our robust partnership with Roche because they have, quite frankly, the feet on the street; the expertise; the competencies, both in the reimbursement and also with the customers, oncologists, and payers and everyone who, in the oncology ecosystem in every market in the world. And so this is -- I'll echo, it's even surprised me a little bit how much our biopharma partners really value -- among many differentiators that we have as a company, but our biopharma partners really appreciate that global footprint. Let's face it, there's no more local studies anymore. They're all global clinical trials.

And to have a consistent tool, a diagnostic that's high-quality, that's identifying the right patients and helping them to hit their endpoints perhaps and have success and identify the right patients at the right time and be convincing with the regulatory authorities, it really puts us in a favorable position to be a biopharma partner of choice.

Patrick Donnelly -- Goldman Sachs -- Vice President

Very helpful, guys. Thanks.

Troy Cox -- Chief Executive Officer

Thanks.

Operator

Our next question comes from Tycho Peterson from J.P.Morgan. Please go ahead with your question.

Tejas Savant -- J.P.Morgan -- Analyst

Hey, guys. It's Tejas on for Tycho. Just one quick one here on the slight CMS delay. Have you had any sort of interactions with CMS over the last couple of weeks or so that would give some indication of what exactly it is that drove the delay? Was it just the volume of comments? And then, in terms of your guidance, are you anticipating any sort of label expansion beyond the five tumor types that you've spoken about in the initial label?

Troy Cox -- Chief Executive Officer

We don't have any additional insights. I mean, we can point to what CMS has communicated publicly. And indeed, they talk about the volume of comments. And as we've communicated, they posted yesterday that they expect to finalize the NCD by March 16.

And that's awfully logical, to write it 60 days after the -- I know it's a bit of a change versus maybe what we had assumed, but a little bit of delay. And so that -- it makes sense. It's 60 days to consider all those comments post the close of the comment period. So not a whole lot more to add other than pointing to what CMS have communicated to everyone.

We do remain encouraged that the final policy will ensure that Medicare beneficiaries with advanced cancer have covered access to well-validated diagnostic tests like FoundationOne CDx.

Jason Ryan -- Chief Financial Officer

And Tejas, related to what we've embedded for NCD assumptions, it's safe to say we haven't gone beyond those five tumor types. I'm comfortable giving you that, and some fairly modest assumptions around CED. We have not assumed, for example, some sort of a tissue-agnostic approval or other indications opened up by CDx getting onto FoundationOne CDx. We focused on the five covered indications.

Tejas Savant -- J.P.Morgan -- Analyst

Got it. And then Troy, can you just share some color on sort of the tenor of your conversations on the private-payer side, particularly given that they manage some of these Medicare Advantage plans? And any updated thoughts on just progress on the NCCN guideline front through 2018? Or is that a largely a black box beyond your control at this point?

Troy Cox -- Chief Executive Officer

Sure. Certainly, and we communicated in prior calls that, obviously, it's a personal focus of mine and also a focus for the company to make sure we have the right efforts with the payers. It's a -- if you just point to a few things, that the evidence is getting more and more compelling. The FDA approval, obviously, helps.

Medicare is 30 -- central Medicare is about 30% of our patients today, and Medicare Advantage is about 10%. And the same payers that we've been working with for private commercial coverage, most of the mid- to large-sized ones are also managing Medicare Advantage. And so the -- it's an opportunity for us to sit at the table with them and discuss this growing compelling evidence and the -- obviously, Medicare coverage, the pull through of Medicare Advantage. And so we've studied and understood exact science and those kinds of approaches to make sure that we can have the smoothest process possible and greatest success, with pulling through the Medicare Advantage will be our top priority, of course, and make sure.

Because that's statutorily obliged by the payers. When central Medicare delivers an NCD, they -- then Medicare Advantage needs to follow suit. Now with that said, we don't expect -- as you've heard me say many times, even with new evidence and ammunition, we don't expect them to fall like dominoes. There'll be a curve like any other curve.

And so we'll -- we're not expecting some silver bullet, all of sudden, a lot of these payers start to cover us. But we certainly think -- and it's reasonable to assume that over time, that we'll make incremental progress with private commercial payers.

Tejas Savant -- J.P.Morgan -- Analyst

Got it. And then one final one for me here, Troy. Can you just remind us about your strategy for China? I know it's probably not going to be very meaningful in terms of the numbers. But back in the day, Foundation Medicine announced a partnership with WuXi.

Obviously now, Roche is handling your OUS sort of commercialization strategy. How should we think about that opportunity? And should we expect any meaningful progress there in 2018?

Troy Cox -- Chief Executive Officer

Yes. What I can say, Tejas, is that we're excited and we're motivated, and it's aligned with our strategy, and equally so with Roche, to continue expansion in all major markets around the world and less than major markets as well, right? So we're going to -- this is a high priority for both organizations. And so it's reasonable to assume that -- obviously, we all recognize the importance of China as a market. Obviously, we're going to pursue every opportunity to deliver our valuable tools around the world and have a great partner to do so.

Tejas Savant -- J.P.Morgan -- Analyst

Got it. Thanks so much, guys. Congrats on the quarter.

Troy Cox -- Chief Executive Officer

Thanks, Tejas. Appreciate it.

Operator

Our next question comes from Tim Evans from Wells Fargo Securities. Please go ahead with your question.

Tim Evans -- Wells Fargo Securities -- Analyst

Just one quick one for Jason on the ASC 606 guidance. Would you be willing to tell us what effect that has in dollar terms on your guidance? Or maybe what the guidance would have been -- revenue guidance would have been under the ASC 605?

Jason Ryan -- Chief Financial Officer

Yes. Tim, it is fairly immaterial effect. Because although you are now beginning to accrue revenue, right, which will have a benefit for, for example, for cases reported in the fourth quarter of this year, you also are not seeing the tail of payments that are coming in from cases reported last year. So the -- what you often see -- what you've seen in the past is that when a company moves from cash to accrual, there's some sort of true-up.

Because you move into accrual right away, but you still have a tail of cash payments, that tail of cash payments for cases unpaid through 2017 all gets swept into an adjustment that gets booked in the first quarter. It's directly to retained earnings and not to revenue. So you're right to sort of initially think that there would be some sort of advantage from moving to the accrual. It is offset in large part because you're not recognizing the cash-based revenue from 2017.

So let me stop and see if that answer made sense to you.

Tim Evans -- Wells Fargo Securities -- Analyst

Yes, it does. I just wanted to doublecheck on that. That's what I figured, but I just wanted to doublecheck. Thank you.

Jason Ryan -- Chief Financial Officer

Great. Sure. Thanks.

Operator

Our next question comes from Raymond Myers from Benchmark. Please go ahead with your question.

Raymond Myers -- The Benchmark Company -- Analyst

Thank you. Let me ask first about the level of milestone revenue that was recognized in the fourth quarter. And how much milestone revenue do you expect in the 2018 guidance?

Jason Ryan -- Chief Financial Officer

This is Jason. Thanks for the question. We have very little milestone revenue in the fourth quarter. Fourth quarter was really driven by -- the pharma side of revenue is driven by profiling and critical trial where we reported over 6,200 specimens to I-O pharma in the fourth quarter.

And so a lot of fourth quarter was not specifically milestone-driven. In fact, even the pharma R&D services line was really not milestone-driven in the fourth quarter. And this is also across a lot of our different partners. With respect to 2018, our milestones will not be accounted for in the way they have been in the past.

It will be much more about percentage of completion as we move through work that then gets paid through a milestone payment. And so we're not breaking out specific milestones because those are not going to show up as revenue the way they have in the past.

Raymond Myers -- The Benchmark Company -- Analyst

Is that related to ASC 606, recognizing it as a percent of completion?

Jason Ryan -- Chief Financial Officer

It is. It is.

Raymond Myers -- The Benchmark Company -- Analyst

OK. Good. That would smooth your revenue over the quarters. Can you give us a sense of how much milestone-related revenue you would expect in the year so that we can compare that to 2017?

Jason Ryan -- Chief Financial Officer

We are really steering away from the milestone element here for a couple of reasons: No. 1, the accounting we just talked about; No. 2 is that as we work now with these 30-plus pharma partners, a number of these agreements include payment that can be considered sort of milestone payment. It's not unusual to have in any one of these contracts elements that are paid as a, quote, milestone.

That is just part of the payment mechanism or the design of the contract. They don't stand out the way we have had, for example, $10 million, $12 million milestones from Roche in the past. One thing I can point you to is that we guided to pharma R&D revenue range of $40 million to $45 million in 2018. Historically, that would be the revenue line where you would see, for example, some of these milestones.

But this year, we're not steering specifically to milestones.

Raymond Myers -- The Benchmark Company -- Analyst

OK, that helps. And next, I wanted to ask about -- generally, the commercial and strategic return that Foundation would expect to get from its large partnerships, for example, Pfizer, BMS, and the other big partnerships, what is the return that you expect to get from those?

Jason Ryan -- Chief Financial Officer

So the return that we expect to get from them is multifactorial. So the first, of course, is the revenue that we're accruing right now. And some of these are multiyear deals. So the revenue we recorded in a period when we closed the deal, of course, is the beginning of a revenue stream.

Second, of course, is the strategic value where we're working with a pharma like a Pfizer, like a BMS, many others across different service offerings and what it can ultimately do for our clinical business. And then some of these do have commercial elements to them. We haven't gotten into specifics of those commercial elements, but we continue to believe we can negotiate some of those effectively over time. But Ray, to specifically get to your question, we don't -- we haven't yet disclosed some of the commercial elements of those pharma agreements.

Raymond Myers -- The Benchmark Company -- Analyst

OK. I can understand that. Lastly, I want to ask a broad question about profitability. You'll be exiting the year presumably with a run -- a revenue of somewhere in the neighborhood of $0.25 billion.

Is it too soon to think about when Foundation might be profitable?

Jason Ryan -- Chief Financial Officer

Well, let me start and I'm Troy may jump in here as well. No, it's never too soon. I think what we have -- what we said is that we look at this really fundamentally in terms of the value that we can bring to patients long term, and we want to maximize that. This a very large, as you well know, a very large opportunity.

Obviously, on the pharma side and certainly on this clinical side in the U.S. and outside the U.S., our job is to maximize this opportunity for patients. So we focus, yes, on operating loss and break-even. We also focus on the right investments to maximize the opportunity.

So we want to ensure that, should we need to, we can move toward break-even. But we don't want to move to break-even too soon and sacrifice expenses or investments to the extent they're going to limit longer-term opportunities. And that's the balance that we have here on an annual and even on a quarterly basis. And so we think we're striking a balance this year through incremental investment.

You can start to see that leverage in the increase in OPEX versus the increase in revenue in the guidance here in 2018.

Troy Cox -- Chief Executive Officer

Yes. And I would add -- it's Troy -- is that we haven't given, and nor are we this time, long-term guidance. But I think those that are close to us appreciate what a unique position we are as one-of-a-kind company and with our unique business model and our approach that keep our leading edge with our biopharma partners. And those partnerships, and fueling their R&D engine, is also fueling our pipeline.

We also have a unique position in terms of the FDA approval and the preliminary NCD. And so it's important for us to invest in these opportunities, and we have strong support from the shareholders that we connect with, the investors, and the board for our financial approach, and feel really good about how we're seizing those opportunities.

Raymond Myers -- The Benchmark Company -- Analyst

Thank you.

Troy Cox -- Chief Executive Officer

Thanks, Rick.

Operator

Once again, if you would like to ask a question please press * and then 1. Our next question comes from Paul Knight from Janney Montgomery. Please go ahead with your question.

Paul Knight -- Janney Montgomery Scott -- Managing Director

Hi, could you give us -- I missed the clinical testing guidance for 2018, Jason?

Jason Ryan -- Chief Financial Officer

Hi Paul. The clinical testing guidance is 90,000 to 100,000 tests. That's about 41% at the midpoint.

Paul Knight -- Janney Montgomery Scott -- Managing Director

OK. And then lastly, with the collaborations you're winning, are you seeing a change in the nature of those collaborations such as attached royalties, higher pricing? What's getting better with these non-Roche collaborations?

Jason Ryan -- Chief Financial Officer

Well, I think what's getting better is what Troy alluded to, is at the breadth of these agreements where we're working with these companies, in many cases, across the portfolio of their products and with a number of our different service offerings. We have not gotten -- I was mentioning to Ray earlier -- we haven't gotten into the commercial specifics, the commercial milestones, if you will. Some of our agreements do have elements of value related to commercial success from pharma partners, and we have not gotten into those details.

Paul Knight -- Janney Montgomery Scott -- Managing Director

And when would you expect -- I think there's a little discussion earlier on the commercial-payer outlook. But what, are you suggesting, what, it's significant in 2019? What's your thought on kind of when you expect it to be significant?

Jason Ryan -- Chief Financial Officer

Yes. I'll reiterate a little bit what Troy said before, which is, we are -- well, we're not guiding to 2019. What we said is that we don't see this overnight. We want to be -- we want to guide appropriately.

While there a lot of moving pieces here, we see NCCN inclusion of broad profiling across a number of different disease states at this point. We see a tremendous amount of data continuing to come out, some from us and some from others. We obviously see the draft NCD, which we anticipate becoming final in Medicare payment. So there are a lot of pieces here, and the Medicare Advantage, last by statute, going under that NCD.

What we don't want to do is guide specifically to commercial third-party-payer timing. We don't see that happening overnight, but we are simply not yet going to get into specifics around quarterly or annual guidance on individual third-party payers.

Paul Knight -- Janney Montgomery Scott -- Managing Director

OK. Thank you.

Troy Cox -- Chief Executive Officer

Sure.

Operator

And ladies and gentlemen, at this time, I'm showing no additional questions. I'd like to turn the conference call back over to Troy Cox for any closing remarks.

Troy Cox -- Chief Executive Officer

Thanks, operator, and thanks to all of you. We appreciate the support of our employees, our customers, and our shareholders and look forward to updating you on our progress in 2018. Thank you for joining us today, and have a great evening.

Operator

Ladies and gentlemen, the conference has now concluded. We do thank you for attending today's presentation. You may now disconnect your lines.

Duration: 62 minutes

Call Participants:

Kim Brown -- Senior Director of Investor Relations

Troy Cox -- Chief Executive Officer

Jason Ryan -- Chief Financial Officer

Amanda Murphy -- William Blair and Company -- Analyst

Doug Schenkel -- Cowen and Company -- Managing Director

Vincent Miller -- Chief Medical Officer

Patrick Donnelly -- Goldman Sachs -- Vice President

Tejas Savant -- J.P.Morgan -- Analyst

Tim Evans -- Wells Fargo Securities -- Analyst

Raymond Myers -- The Benchmark Company -- Analyst

Paul Knight -- Janney Montgomery Scott -- Managing Director

More FMI analysis

This article is a transcript of this conference call produced for The Motley Fool. While we strive for our Foolish Best, there may be errors, omissions, or inaccuracies in this transcript. As with all our articles, The Motley Fool does not assume any responsibility for your use of this content, and we strongly encourage you to do your own research, including listening to the call yourself and reading the company's SEC filings. Please see our Terms and Conditions for additional details, including our Obligatory Capitalized Disclaimers of Liability.

More From The Motley Fool

• 3 Growth Stocks at Deep-Value Prices

• 5 Expected Social Security Changes in 2018

• 6 Years Later, 6 Charts That Show How Far Apple, Inc. Has Come Since Steve Jobs' Passing

• 10 Best Stocks to Buy Today

• The $16,122 Social Security Bonus You Cannot Afford to Miss

• Bitcoin's Biggest Competitor Isn't Ethereum -- It's This

The Motley Fool has no position in any of the stocks mentioned. The Motley Fool has a disclosure policy.