Castle Biosciences Inc (CSTL) 2020 Q4 法說會逐字稿

Good afternoon, and welcome to Castle Biosciences Fourth Quarter and Full Year 2020 Conference Call. As a reminder, today's call is being recorded. (Operator Instructions) I would now like to turn the call over to Frank Stokes, Chief Financial Officer. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to Castle Biosciences Fourth Quarter and Full Year 2020 Financial Results Conference Call. Joining me today is Castle's Founder, President and Chief Executive Officer, Derek Maetzold.

Information recorded on this call speaks only as of today, March 8, 2021. Therefore, if you are listening to a replay or reading the transcript of this call, any time-sensitive information may no longer be accurate. A recording of today's call will be available on the Investor Relations page on the company's website for approximately 3 weeks.

Before we begin, I would like to remind you that some of the information discussed today may contain projections or other forward-looking statements regarding future events or the future financial performance of the company, including expectations and assumptions related to the impact of the COVID-19 pandemic, and are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. The forward-looking statements are based upon current expectations and involve inherent risks and uncertainties, and there can be no assurances that the results contemplated in these statements will be realized.

A number of factors and risks could cause actual results to differ materially from those contained in these forward-looking statements. These factors and other risks and uncertainties are described in detail in the company's annual report on Form 10-K for the year ended December 31, 2020, and in the company's other documents and reports filed with the Securities and Exchange Commission. These forward-looking statements speak only as of today, and we assume no obligation to update or revise these forward-looking statements as circumstances change.

I'll now turn the call over to Derek.

Thank you, Frank, and good afternoon, everyone. Thank you for joining us today. This afternoon, I will discuss highlights from our significant achievements during 2020, recap select fourth quarter results, provide thoughts on current trends and then close on why we remain confident in our ability to create near- and long-term growth. And then Frank will provide additional detail on our financial performance.

But before we get started this morning, I would like to discuss the announcement we made today regarding the launch of our DecisionDx-Melanoma integrated test result. We believe this launch represents a significant step forward in the care of patients diagnosed with cutaneous melanoma.

Specifically, DecisionDx-Melanoma now outputs an integrated test result calculated by an independently validated artificial intelligence-based algorithm, what we call i31-GEP, to provide a more precise prediction of sentinel lymph node positivity in order to guide discussions and recommendations for the sentinel lymph node biopsy surgical procedure.

The i31-GEP test report is based on artificial intelligence-based neural network algorithm discovered and developed or trained on a multi-center cohort of 1,398 patients locked and then independently validated in a separate cohort of 1,674 prospective consecutively tested patients with melanomas of all thicknesses. The i31-GEP test report integrates an individual patient's traditional clinical and pathologic features with the DecisionDx-Melanoma test result score.

What's exciting here is that based upon linear regression slope analysis, the integrated test result provides a near-perfect relationship between the predicted and the actual observed sentinel lymph node positivity outcome in our validation cohort.

We also expect to launch a mobile and web-based application on the i31-GEP test later this year. The app is designed to allow clinicians to have an interactive platform to discuss their patient's risk on the integration of clinical and pathologic features with the DecisionDx-Melanoma test score. We are excited about this evolution of our DecisionDx-Melanoma test.

So what does this advance to precision with an integrated test result mean on a practical level? Let's take a specific patient example. A 62-year-old male patient with a cutaneous melanoma diagnosis, and that melanoma has a tumor thickness or a Breslow's depth of 0.7 millimeters. Using population-based statistics from AJCC, this individual patient would have had a likelihood of a positive sentinel lymph node biopsy result of less than 5% using those clinical and pathologic features alone. That's important, because the NCCN guidelines do not recommend that the sentinel lymph node biopsy procedure be offered in a patient with a predicted less than 5% sentinel lymph node positivity.

Now if we take that same patient and he has tested with our DecisionDx-Melanoma test and receives the lowest risk score or risk class, a Class 1A test result, which has an accompanying DecisionDx-Melanoma score of 0.2, and that's out of a range of 0.0 to 1.0, then our integrated test result would predict the likelihood of sentinel lymph node positivity of 3.6%, less than the 5% predicted by clinical and pathologic features alone.

However, now let's look at that same 62-year-old male patient, but now his DecisionDx-Melanoma test result is our highest risk class, Class 2B, with an accompanying score of 0.8. Now the same patient, but with a different tumor biology profile with the integrated test result, is predicted to have a positive sentinel lymph node likelihood of 15.6% compared to the 3.6% likelihood with the Class A score I just mentioned and compared to the clinical and pathologic features, population-based risk of less than 5%.

Again, without having the benefit of our DecisionDx-Melanoma test, based upon clinical and pathologic features alone, this patient would have been managed as if they had less than a 5% likelihood of known positivity. And under NCCN guidelines, with that positivity number, they would not have been offered a sentinel lymph node biopsy procedure.

However, by using our DecisionDx-Melanoma test in this patient, they had a predicted sentinel lymph node positivity of 15.6%, which is significantly above a 10% threshold that NCCN uses for recommending that a sentinel lymph node biopsy procedure be offered to that patient. This is clearly a more precise finding that will have a positive impact on patient care.

So today's launch of our DecisionDx-Melanoma, integrated test result is designed to provide more precise, more personalized prediction of the risk of sentinel lymph node positivity in order to allow patients to move from population-based to personalized and into more precision-based medicine and precision-based care. We are planning to move our other tests in this same direction, both the commercially available tests and our tests in development. These are exciting outcomes at Castle Biosciences.

Now as always, I would like to thank the Castle team. I cannot express how proud and thankful I am with what we achieved in 2020 and the resiliency shown throughout the organization. Our team's hard work and commitment to improving the lives of patients with skin cancer allowed us to successfully navigate through this unprecedented year. We remain focused on our strategic initiatives and believe we are well positioned for future growth, providing dermatologic clinicians with valuable genomic information to better inform their treatment decisions.

In 2020, we delivered revenue growth of 21% over 2019, and our dermatologic gene expression profile test report volume, which includes our 2 new tests launched in the second half of 2020, grew over last year by 8%. This growth occurred despite a more than 20% reduction in melanoma diagnoses in 2020 compared to 2019, which I will discuss further in a moment.

Key components of our near- and long-term growth strategy include advancement of our innovative pipeline tests, including the 2 launches in the second half of 2020, as well as moving toward precision medicine that I just described, evidence developments, reimbursement gains and the initiation of new pipeline programs. We achieved all of our 2020 milestones in these areas.

We started 2020 with 1 skin cancer test and ended the year with a suite of 3 tests, more than tripling our estimated end market U.S. total addressable market to $2 billion. This is a significant achievement for any year, even more so as we face extraordinary challenges from COVID-19.

You may recall that Castle's proprietary DecisionDx-Melanoma test for patients diagnosed with cutaneous melanoma was developed and validated in-house. With the launch of our 2 additional skin cancer tests, DecisionDx-SCC for patients diagnosed with 1 or more high-risk features with cutaneous squamous cell carcinoma and DecisionDx DiffDx-Melanoma designed to aid in characterizing difficult-to-diagnose melanocytic lesions, we further demonstrated our ability to identify dermatologic diseases with high unmet clinical need where genomic tests have the potential to improve patient management decisions and then brought these tests to market.

In 2020, we expanded our significant body of evidence with 11 supportive peer-reviewed articles published for our 4 proprietary gene expression profile tests and expect additional publications in 2021, having already received acceptance notifications. We believe our existing body of evidence provides us with a significant competitive advantage over potential entrants into the U.S. market.

DecisionDx-Melanoma alone is supported by 28 peer-reviewed publications, and through December 31, 2020, has been ordered close to 69,000 times for use in patients with cutaneous melanoma. We expect to continue to invest in evidence development, as it remains a key component of our growth strategy, supporting adoption of our tests by clinicians and reimbursement by commercial payors.

Turning to 2021. Although we are seeing areas of recovery, due to COVID-19, we continue to experience some headwinds to our GEP test support volume growth rate, including DecisionDx-Melanoma, our current lead revenue driver, which we believe is a result of reduced diagnoses of cutaneous melanoma. As I previously discussed, 2020 diagnosis of melanoma were down by more than 20% or approximately 26,000 diagnoses compared to 2019. We continue to believe this is due to a mix of temporary closing of dermatology practices and/or patient volume in the second and third quarter of 2020 as well as the shifting from in-person general health care business to telemedicine visits, which we believe was a significant factor in the ongoing reduction in the diagnosis of melanoma, especially in the back half of 2020 and early 2021.

For example, typically, a patient would go in for their regular high blood pressure, diabetes or just an annual physical visit, and the primary care clinician would identify a concerning mole, which after in-person testing will be identified as a melanoma. However, with telemedicine business, we believe these identifications are not taking place as often. Given that our DecisionDx-Melanoma test is ordered by physicians after biopsies take place and the diagnosis of melanoma is rendered, a delay in the diagnosis of melanoma does reduce the pool of eligible patients.

In addition to the impact of COVID, we believe the severe weather we have experienced in the first quarter in many parts of the U.S. has impacted dermatology practices and patient flow to date. As you may recall, historically, our first quarter is generally flat to the fourth quarter. Therefore, given the continued impacts from COVID, recent weather and historical first quarter seasonality, we expect our first quarter 2021 DecisionDx-Melanoma test result -- report volume to be softer than the fourth quarter of 2020. However, due to the effect of our first full quarter of expanded Medicare LCD for DecisionDx-Melanoma and other factors for the first quarter of 2021, we don't expect the sequential softening in report volume over the fourth quarter of 2020 to translate into a commensurate softening in revenues in the first quarter of 2021 compared to the fourth quarter of 2020.

Although we can't predict the trajectory of the recovery of these missing diagnoses of melanoma, we anticipate improvement in our volume growth rate for the second half of '21 and into 2022, although we expect that it will not be linear. Specifically, we expect the majority of the estimated 26,000 patients, who were not diagnosed in 2020 as well as those who are not diagnosed in early 2021, to be diagnosed later in '21 or in '22, which could be impacted by the timing of both COVID vaccinations and the return to in-person medical care. We know that the best patient outcomes are linked to early detection of cancer. So for the best patient care and outcomes, we are hopeful that these diagnoses will happen sooner rather than later.

As you know, one important metric for us is the achievement of reimbursement milestones with our proprietary innovative tests. Our DecisionDx-UM for UV melanoma and DecisionDx-Melanoma tests have both been designated by CMS as ADLTs or Advanced Diagnostic Laboratory Tests. Practically, this means that the Medicare rate for our tests are evaluated every year based upon the median allowable private payor rate. For 2021, our rate for DecisionDx-UM was set at $7,776, and our rate for DecisionDx-Melanoma is $7,193.

In addition to our 2021 Medicare rate, it's important to remind you of the expansion of the LCD for DecisionDx-Melanoma. You may recall that we received initial LCD coverage in late 2018. The initial LCD covered the use of our DecisionDx-Melanoma test to guide sentinel lymph node biopsy surgical decisions. Palmetto and Noridian at that time did not evaluate coverage for the other management decisions such as initiation of imaging surveillance, follow-up frequency, referrals to oncology, et cetera, that are in clinical use today.

The initial LCD covered around 45% of clinically tested Medicare patients. We subsequently worked with Palmetto and Noridian regarding expansion of the initial coverage, as many clinicians were using our DecisionDx-Melanoma tests for those other management decisions.

Subsequently, an expanded draft LCD was posted in 2019 that covered those additional uses. This draft was converted to a final LCD that was effective on December 6, 2020. We estimate that the current LCD now covers slightly more than 90% of all clinical uses. Meaning that in 2021, we expect the DecisionDx-Melanoma test to be a covered benefit for more than 90% of all Medicare beneficiaries.

Let's now turn to the 2 skin cancer tests we launched in the second half of 2020. The technical assessment dossier for DecisionDx-SCC test was submitted to Palmetto and Noridian in the second quarter of 2020. We received confirmation of acceptance of the submission as being complete in the third quarter 2020 and believe, although there can be no assurances, that a draft LCD should be posted in 2021. We believe the LCD could be posted in the first half of 2021, but I remind you that there is no specific time frame under which Palmetto and Noridian must operate.

We have also, in early 2021, submitted our technical assessment dossier for our DecisionDx DiffDx-Melanoma test. We are hopeful that the time frame for review for our DiffDx-Melanoma test will be shortened as Palmetto and Noridian had previously reviewed Myriad's myPath Melanoma test and found the intended clinical use to meet the definitions of medical reasonableness and necessity. As a reminder, our DiffDx-Melanoma test has the same intended clinical use as Myriad's test. Accordingly, we also expect that a draft local coverage determination policy should be posted in 2021 with finalization in 2022.

Now I'll briefly recap 2020 revenue and test volume highlights. In 2020, we increased revenue by 21% to $62.6 million from $51.9 million in 2019. This includes positive revenues related to prior periods of $0.2 million for 2020 compared to $2.5 million for 2019. As our business continues to evolve, we are focusing on total dermatologic test volume as a key metric for measuring our success.

As a reminder, we announced in January that we delivered 16,790 total dermatologic gene expression profile tests in 2020, which represents an 8% increase in total reported volume for our dermatologic tests compared to 2019. Specifically, DecisionDx-Melanoma test results delivered in 2020 were 16,232. DecisionDx-SCC test reports delivered in 2020, since the August 31 launch, were 485. And the DiffDx-Melanoma test reports delivered in 2020, since they were launched on November 2, 2020, were 73.

Initial response to our 2 new skin cancer tests, DecisionDx-SCC and DiffDx-Melanoma, did exceed our expectations. We believe this early performance is due in part to the high unmet clinical need that our tests are filling. Equally important, though, is the value that we are seeing through leveraging our dermatologic commercial channels. Being able to walk into an office of a current customer, one who has adopted DecisionDx-Melanoma as an important tool in patient management, makes it easier to lay out the clinical need and value of our DecisionDx-SCC test.

In 2020, we successfully executed on 2 capital raises that together generated approximately $330 million in net proceeds. We plan to increase the investments we are making in the company to facilitate our growth initiatives and remain in a position of strength in the near term as well as lay a foundation for future growth. These initiatives include the expansion of our commercial team, acceleration of our R&D efforts to support our existing tests and the advancements of our dermatologic pipeline tests.

Regarding the expansion of our commercial team, you may recall, we successfully completed 2 commercial expansions in 2019. We went from 14 outside sales territories to 23 and then to 32 by the end of 2019. Then in the third quarter of 2020, we added 10 additional sales representatives dedicated to the DiffDx-Melanoma test, our test for suspicious pigmented lesions, with their focus being to introduce the DiffDx-Melanoma test to dermatopathologists during the first 6 months of clinical availability before we expand introduction into the clinical dermatology community.

We are actively recruiting for additional 15 to 20 outside sales territories. Our goal is to exit the first half of 2021 with approximately 60 outside sales reps, all of whom will sell our entire suite of skin cancer tests with call points focused on dermatologists; Mohs surgeons; surgeons who work in skin cancer, including surgical oncologists and head and neck surgeons; and dermatopathologists. We have seen that our GEP test report volume growth is responsive to promotional efforts, and we are excited to see continued volume growth of our skin cancer test franchise.

We are also accelerating our research and development activities, including 2 important studies that support our DecisionDx-Melanoma test. As previously discussed, we initiated 2 significant protocols for our DecisionDx-Melanoma test in 2020, and both are progressing well.

The first is the personalized study in which we will evaluate DecisionDx-Melanoma for interactions with adjuvant therapies. The second is a connection study which will collect long-term outcomes for up to 10,000 patients who have been tested with DecisionDx-Melanoma. Together, these studies will add valuable data to our existing biobank of data on patients with melanoma and their tumor biology, which consists of approximately 60,000 tissue samples to date. We believe this is the largest database of its kind and are exploring opportunities to leverage the value of these data.

With our industry-leading position as the only diagnostic company with 3 commercially available tests for dermatologic cancer, we have demonstrated our ability to successfully develop, validate and bring to market clinically actionable, innovative genomic tests. We start by identifying dermatologic diseases with high unmet clinical need, and we use the gene expression profile of the individual patients tumor biology to assist clinicians and their patients better informing treatment to optimize health outcomes and reduce health care costs.

Due to the biological complexity of dermatologic cancers, developing accurate products take scientific diligence, stringent clinical protocols, artificial intelligence expertise, proprietary algorithms and significant investments of time and capital.

Building on our proven process, I am pleased to announce that we have initiated working on a number of pipeline tests, which branch out upstream, downstream and parallel to our currently marketed tests. Although we are still in the early stages of development, we believe that we can launch 3 to 5 new tests by the end of 2025 utilizing our existing sales channels, potentially adding approximately $3.6 billion to our current U.S. total addressable market, with the result being a U.S.-only TAM of slightly more than $5.5 billion.

As I said, these tests are in early stage development, but we expect to provide additional details as soon as appropriate and before the end of 2021. Additionally, the leverage we see on the commercial side of our business translates to our clinical research initiatives. We had 220 active clinical research sites or centers in 2020. We anticipate we will work with many of these same centers on our development initiatives as those who worked on with the DecisionDx-Melanoma, DecisionDx-SCC and DiffDx-Melanoma.

I will now turn the call back over to Frank, who will provide additional detail relating to our financial results.

Thank you, Derek. We're pleased to report solid financial results, and along with the significant progress we made on our growth initiatives, we believe we are well positioned for near- and long-term growth.

We reported revenue of $17.3 million in the fourth quarter of 2020 compared to $17.6 million in the fourth quarter of 2019, which included $3.5 million in positive revenues related to prior periods compared to $4.3 million of such revenues in the fourth quarter of 2019.

Our full year 2020 revenue was $62.6 million, an increase of 21% over 2019, primarily due to higher per-unit revenues and increased overall test volume, partially offset by reduced positive revenue adjustments related to prior periods. Full year 2020 revenue includes positive adjustments related to tests delivered in prior periods of $0.2 million compared to $2.5 million for 2019.

We were able to successfully scale the organization in 2020, achieving our internal expansion goals and headcount moved from 135 on December 31, 2019, to 201 as of December 31, 2020, representing growth of close to 50%. And in 2021, as of March 2, we had 224 employees and continue to fill open positions.

In 2020, we maintained strong gross margins, and for the fourth quarter and full year both, our gross margin was 85% compared to 89% and 86% for the fourth quarter and full year of 2019, respectively, reflecting the expansion of laboratory staff in preparation for both launches of DecisionDx-SCC and DiffDx-Melanoma as well as in preparation of anticipated further volume growth for DecisionDx-Melanoma.

As a reminder, a significant portion of our cost of sales expense represents fixed costs associated with our testing operations. Accordingly, our cost of sales expense will not necessarily increase or decrease commensurately with the change in net revenue from period-to-period.

Our operating expenses for the fourth quarter and year ended December 31, 2020, were $17.7 million and $59.5 million, respectively, compared to $13 million and $37.2 million for the comparable 2019 periods. The increase was driven primarily by higher SG&A, which increased by $5.1 million for the quarter and $18.3 million for the year, attributable in part to the expansion of our sales and marketing teams for the launch of our DiffDx-Melanoma test, administrative support functions and higher personnel costs associated with our increased headcount, which includes salaries, bonuses, benefits and stock-based compensation.

R&D expense increased by $5.9 million in 2020 compared to 2019 and was primarily associated with increases in personnel costs attributable to additional headcount and costs incurred in our clinical study. As Derek discussed earlier, the acceleration of R&D efforts is a key growth initiative, and we expect to further increase our R&D expense as we continue to fill critical roles, progress key clinical studies, including the 2 Derek just mentioned, and continue to invest in activities that support our products, pipeline investment and position us well for continued growth.

Our recognition of other operating income of $1.9 million in the fourth quarter of 2020 is attributable to a favorable change in requirements associated with the provider relief funds program under the CARES Act.

Interest expense decreased $1.9 million in 2020 compared to 2019, primarily due to interest on the convertible promissory notes that were outstanding in 2019. These notes converted into common stock in connection with the IPO in July of 2019. The decrease also reflects lower interest expense on our banking term loan facility, primarily due to lower average balances outstanding. We terminated and repaid all amounts due on the term loan facility in December 2020 and currently have no debt. We recorded an extinguishment loss of $1.4 million during the year ended December 31, 2020, related to the early repayment and termination of our banking term loan facility.

2019 included certain nonoperating items primarily related to debt conversion and mark-to-market activity leading up to the IPO. These items had a net positive effect of income on $2.3 million for the year ended December 31, 2019.

Noncash stock-based compensation expense, which is allocated among cost of sales, R&D and SG&A, totaled $8.3 million for the year ended December 31, 2020, compared to $1.2 million for the year ended December 31, 2019. We expect material increases in stock-based compensation expense in future periods, reflecting both higher post-IPO stock option valuations as well as additional awards outstanding due to growth in our headcount.

Our net loss for the 12 months ended December 31, 2020, was $10.3 million compared to net income of $5.3 million for the 12 months ended December 31, 2019. Diluted loss per share attributable to common stockholders for 2020 was $0.54 a share compared to diluted loss per share attributable to common stockholders of $0.21 for 2019.

Operating cash flow for the 3 months ended December 31, 2020, was negative $0.4 million compared to positive $4.5 million for the same period in 2019. Adjusted operating cash flow, a non-GAAP measure, for the 3 months ended December 31, 2020, was $1.5 million. For the 12 months ended December 31, 2020, we generated $9.9 million of operating cash flow compared to $7 million during the same period in 2019. As a reminder, our operating cash flow for the full year 2020 benefited from an advanced payment of $8.3 million from CMS, which will be applied against future Medicare claims that we submit for reimbursement.

Recruitment will begin in April 2021 and continue for a period of up to 17 months. We have excluded receipt of the advanced payments from adjusted operating cash flow, but as future claims are submitted for reimbursement and applied against this balance, we expect to include the advanced payment and adjusted operating cash flow as it's recouped.

Excluding the $8.3 million payment, our adjusted operating cash flow for the 12 months ended December 31, 2020, was $1.5 million compared to $7 million for the same period in 2019. Cash used in investing activities was $4.8 million in 2020 compared to $0.9 million in 2019 due to additional capital expenditures primarily related to expansion of our facilities. Net cash provided by financing activities was $305.9 million in 2020, primarily reflecting proceeds from the 2 public offerings of common stock, partially offset by debt repayments, compared to $88.3 million cash provided by financing in 2019, the majority of which was attributable to the IPO.

Finally, we ended 2020 with a fortified balance sheet. We had cash and cash equivalents at December 31, 2020, of $410 million and 0 debt. As we continue to make progress, positioning ourselves as the leader in the dermatologic genomic testing, our near- and long-term capital allocation priorities remain unchanged and we believe allow us to continue creating shareholder value. They include: first, acceleration of our R&D efforts to build our expansive body of evidence that supports our marketed tests as well as to develop our robust pipeline of tests; and second, the continued expansion of our sales and marketing team.

Turning to our 2021 outlook. Due to the uncertainties regarding COVID-19 and the related impact on the diagnoses of melanoma, we are not providing formal revenue guidance at this time. Although we can't predict the trajectory of any recovery, our core business and underlying fundamentals remain strong. We are excited about the opportunities that lie ahead and remain committed to building a dermatologic diagnostics company focused on making a positive impact on patient outcomes.

I'll now turn the call back to Derek.

Thank you, Frank. In summary, our team's resilience and strength allowed us to deliver strong results and achieve all of our milestones for 2020, despite the unique challenges we all faced. As we continue to build our dermatologic franchise in '21 and beyond, we will continue to put the patient at the center of everything we do.

Before we move on to Q&A, I want to again express my gratitude to our employees. They drive our success. We tend to rely on them to continue to innovate, execute and focus on improving the lives of patients diagnosed with skin cancer.

This concludes our remarks. Thank you for your continued interest in Castle. Operator, we are now ready for Q&A.

(Operator Instructions) Our first question comes from the line of Puneet Souda from SVB Leerink.

So first one is on the guide. I appreciate you pointing out softness in the first quarter in cutaneous melanoma volumes, given the pandemic. But just wondering, in terms of the step down here, should that be in line with what the step-down we have seen -- low to mid-single-digit step-down that we have seen here in third quarter to the fourth quarter? Is that the right magnitude to think about it?

And then just broadly, when we think about the full year, you obviously have a number of growth drivers here in commercial sales force that's expanded -- headcounts in commercial that's expanded significantly and vaccinations are ongoing. So sort of maybe if you can provide us maybe what penetration in terms of cutaneous melanoma should we be expecting for the year, if you can provide us the volume for the year?

Thanks, Puneet. Just to clarify, when you're talking about step-down, are you talking about volume sequentially just through the quarter or -- I want to make sure I understand what you're looking for?

Yes. Sequentially for the first quarter, yes.

Yes. So recall, Puneet, typically, fourth quarter to first quarter is sort of flattish. There's a number of factors that drive that. We think we understand what most of them are -- I mean I think a lot of them are. There's an impact of patients' deductibles resetting in January. And so you take the December holidays in Q4. You take that dynamic in Q1. We typically see flattish performance. When you go back and look year-over-year, the sequential quarter performance looks flattish.

This year, we've got continued COVID impact, and we've also got some weather dynamics this quarter. So although it's early yet and we aren't able to quantify, it does look like we're going to see some sequential softening in volumes just based on the continued macro factors facing physician visits.

Now as it relates to penetration, the difficulty there is just trying to predict how many melanoma diagnoses there are going to be this year. Having said that, we're going to very soon have a much larger megaphone for the physician audience. And we have been optimistic that we are timing that with a corresponding resumption of normal commerce as we get through this vaccine cycle and get through some of the broader reopenings.

And as we've said before, a year ago, I think we all sort of felt like maybe a switch would get flipped and everything would go back to normal uniformly. But we're clearly seeing regional differences, and we're seeing differences within the regions. So I think getting a normalization of commerce will get back to that normal rate of diagnoses, and then we can be a little bit more predictive in terms of what penetration of growth is going to look like.

That being said, as we noted, Puneet, despite what doesn't seem to be a recovery from the normal rate of diagnosis of melanoma in the first quarter, at least from third-party data, we aren't seeing a commensurate softening in revenues at this point in time. So I would delink those 2 slightly.

But the other question here, which is when of those estimated 26,000 patients, who were not diagnosed last year, begin showing up and I think the worst thing we can do from a public standpoint is to aggressively assume they all come back quickly. But those melanomas aren't going away. They're going to be diagnosed at some point in time. I don't know if that really starts at the beginning of third quarter of this year, a little bit in the second quarter or fourth. But certainly, you would think of the next 1.5 years, you would see a return to a normal cadence of the diagnosis of melanoma from a year-over-year standpoint.

Okay. And then Frank, in terms of the prior period, we obviously saw a pickup here. How should we think about that prior period quarters going forward? Because I think the expectation initially was, as the penetration rises as -- the prior period revenue number should be declining over time. So maybe just give us a view as to what -- how you're viewing revenue on that line?

Yes. Puneet, I think we're going to continue to see the prior period number come down as we get more and more accurate in our accrual rates for our cutaneous melanoma business. Now having said that, that's going to be offset somewhat by the new products that are being launched. Because the revenue there will be -- well, almost all of that will be prior period as we collect on that just due to the appeal cycle. So it should converge us on the CM business. And then if we are quite successful on the launch products, then you could see that continue to be a factor in the revenue line.

Okay. And then my last one is on the sales reps. Can you maybe give us a sense of mix of the sales reps that are doing face-to-face detailing with the docs today versus those that are online? Or maybe how much of the time is face-to-face versus online today? And obviously, expectation here would be that they would -- that would turn face-to-face in the second half of the year.

Yes. So I'll come on the last half first. So I think you're right. I think our expectation is that -- I don't know what the percentage of face-to-face visits versus, say, having a really engaged conversation over the phone or e-mail and text was in 2019, as we didn't quite capture data that way. But I would think that we would see, based upon current trajectories, a relative beginning of normalcy in kind of third and fourth quarter, as you said.

We haven't pulled data for the first quarter yet, but we just came off of 2 semi-national sales meetings the last 2 weeks. And it seems to me, as Frank kind of mentioned earlier, it's regional and basis right now. You see a strong opening in areas where there's a lower level of case numbers and maybe more of an opening up of the society there, as vaccines roll the way through, health care providers are largely taking care, of course.

And so our representatives are doing more and more vaccine or post-vaccine environment, we'd expect that to open up slowly, I would think, in the second quarter. And as you've kind of indicated, all the macro press seems to assume that we'll be back into sort of a closer normalcy in the second half of 2021.

Our next question comes from the line of Paul Knight from KeyBanc.

The DecisionDx test, obviously, on squamous getting some traction. Can you talk about what your reception in the market was? And obviously, a nice sequential bump there and what you're thinking about as Q1 and the year rolls out?

Paul, are you thinking about the squamous cell carcinoma test specifically?

Yes.

Yes. Yes. I think the reception of the squamous cell carcinoma test has been higher than our expectations where from kind of a forecast standpoint. And that's despite having us really sell the product or introduce it behind the DecisionDx-Melanoma test. So even though it's a launch product, we actually tried to orient the field forces to basically continue to go ahead and talk first about the melanoma test, and then secondly as there's time, to squamous cell test. So even though that dynamic was there, we were quite pleased with the uptake, and again, ahead of our internal estimates.

So why is that? I think that there's a mixture of 2 things, and I don't know which is more important, to be honest at this point in time. But clearly, an important factor is that there's such a great unmet need for clinicians and their patients to make a shared decision about the next course of action when you've been diagnosed with a squamous cell carcinoma of the skin that has 1 or more high-risk features, because the accuracy of just pathology staging alone is just not that great.

So I think one is, we identified several years ago with our -- actually our current melanoma customers a significant unmet clinical need. And I think the test is meeting the expectations clinically.

I think the other part of it, though, again, which I don't know, which is more important, is the fact that the -- we anticipate that the ordering clinician, the ordering customer for a squamous cell carcinoma test is likely to have, what, 90% overlap at the end of the day with current melanoma customers. So I think the fact that these dermatologists, the Mohs surgeons and NPs and PAs know the company. If they're ordering our melanoma test, they hopefully have thought through how to incorporate the results of a DecisionDx-Melanoma test and the patient care.

We walk in and that reputation, that expectation that I know how to use the test translates over once he starts talking about squamous cell carcinoma. So again, I don't know which one of those 2 factors is more important, except that certainly if there was no need clinically, you shouldn't expect any test orders to come in the door. But certainly, having a, hopefully, what I believe is, a very positive reputation among the melanoma customers, I think, translates pretty quickly to, okay, I get -- I can see the need. Now let's talk about the patient population that could benefit.

And then can you talk a little bit about the competitive environment, alternative tests out there? Our impression is it's -- probably you're in a better position than you were based on our competitor review, but any thoughts there?

For the squamous cell test again?

For melanoma specifically.

On the DecisionDx-Melanoma test, so we don't believe we have, at this point in time, any direct competitors in the U.S. offering tests today. We do anticipate having 1 or 1 of 2 of the smaller private European companies might be able to launch their test later this year, maybe next year. But as of now, the real competitor is really standard of care and helping clinicians appreciate that after 26 peer-reviewed publications, there's a significant weight of evidence that supports the use and value of our test and move them from point A to point B.

Now it's interesting, pre-COVID, we had staffed up in December of 19 to 23 sales representatives -- I'm sorry, 32 from 23. And they were trained largely in December, and we sort of, therefore, had kind of a 50% expansion in the first quarter of 2020 ahead of COVID really impacting medical commerce. And we saw significant promotion responsiveness. I believe that as we scale our organization from 32 dermatology-facing sales representatives to just above 60 in the next couple of months that will enter into the second half of the year in the same kind of response position that we saw a year ago this time.

Our next question comes from the line of Sung Ji Nam from BTIG.

Just a couple of quick ones. One, on the integrated test results, obviously, that sounds very intriguing. Would you be able to talk about -- I'm not sure if it's too early, but what's the magnitude of improvement adding that component to the overall test results from a performance characteristic standpoint? I don't know if that's the right way to think about it.

Yes. Let's see here. So I don't have an answer that's going to be quite that Direct, Sung Ji, sorry about that. If I think about sort of using clinical pathologic features alone for clinicians to decide who to rule in and who to rule out for a sentinel lymph node biopsy procedure. If you then incorporate using continuous variables, wherever you can find them, our test itself, Breslow thickness, mytotic rate, et cetera, we can reshuffle those patients, both upwards and downwards. And that does result in the majority of patients being able to fall below a 5% likelihood of sentinel lymph node positivity.

So that's a significant reduction over what you would have with clinical and pathologic features alone. As I mentioned in that example that we talked about on the conference call, we also can do the opposite now with much more precision. So we can take somebody, who falls below the sort of threshold for thinking about performing a sentinel lymph node biopsy procedure. And by using this integrated test result score, we could actually find people who would be recommended for sentinel lymph node biopsy consultation, should you decide to incorporate the tumor biology on top of pathology.

So that to me is a tremendous opportunity to both rule out, where appropriate, with very, very high precision and also rule in at the same time.

Got you. That's helpful. And then just on the -- obviously, seeing very good initial volumes for squamous cell carcinoma, the SCC test. And I know you guys are targeting your existing DecisionDx-Melanoma customers. But are there cases where -- I don't know if it's again too early to tell, are there cases where some of the early adopters of SCC you might be able to turn them into more robust users of DecisionDx-Melanoma? Or even as we think about kind of down the road, could that also be tried?

I think the answer is yes. So right now, there are probably several hundred physicians who are Mohs surgeons, who are not ordering clinicians of our DecisionDx-Melanoma test. So why is that? Could be access point, lack of promotion response to this, because we haven't been able to promote to them. Could be because they are inside of a larger dermatology practice and they do mainly Mohs surgery full time and don't really think about the melanoma patient they might be doing excision work on, but think maybe that's a medical dermatology partner orders our melanoma test.

We go and talk to them about the squamous cell carcinoma test, and in most cases, the Mohs surgeon is the one who's seeing these patients with 1 or more high-risk factors or features. We would hope that we would see the same kind of back benefit in terms of having them assess our squamous cell test and hopefully then also be open and be listening to the melanoma test in terms of where they could actually take charge of that from an ordering standpoint.

So we do expect both the dermatologists who uses our melanoma test to be the adopter of our squamous cell test. And we expect to see a minority of clinicians, probably mainly Mohs surgeons who would begin to be interested and hopefully adopt our squamous cell test to get back into seeing the clinical utility of our melanoma test. So I think both ways benefit us going forward.

Our next question comes from the line of Catherine Schulte from Baird.

I guess, first, you've talked about the sales force incentives for SCC being focused on ordering clinicians rather than volumes. Can you just comment on what kind of yardsticks are being measured against? Could 30% of your existing derm customers be ordering SCC by the end of 2021? Or just how would you judge success there in terms of ordering clinicians by the end of the year?

It sounds like a more than less answer. Do you want to comment on that?

I don't know -- Catherine, I don't know if it's -- I don't know if we can give precision in terms of target for numbers. But we -- what's interesting and what's exciting is that when you have a position that embraces and understands the value of gene expression profile testing for one disease state, it's quite easy to get them to see that value in another. And so while we have certainly seen the success of melanoma make the launch of squamous cell more exciting, I think we're going to see the reverse improvement as well.

I think you're going to have physicians who order squamous cell first who say, I see this value, and by the way, I've also got melanoma patients and I want the same value there. So I guess, aspirationally, I think all of the physicians should order both, Catherine, but that's probably a bit hopeful. But I do think there will be -- continue to be significant overlap. And as we said, most of the orders for squamous came from physicians who are users of melanoma. So I think we'll continue to see those 2 numbers grow fairly in lockstep, I would think.

Yes. I don't know what the specific target bonuses are set up in terms of number of new ordering doctors for SCC in 2021, just to be frank, I guess, you should say you should know that number, but I don't. But I think our expectation is that of the 4,500 clinicians who roughly ordered our tests in 2020 -- I guess, that was actually third quarter data maybe. Anybody who's seeing invasive cutaneous melanoma is likely to also be seeing squamous cell carcinoma probably 2.5 to 3, 4x that number. So it would be very surprising that at the end of the day, not the end of 21 months, at the end of the day, the majority of those DecisionDx-Melanoma customers are not also going to be adopting SCC.

I think in a sort of funny period of coming out of limited face-to-face meetings in the fourth -- in the first quarter of this year, probably tailing in the second quarter. What that really means for kind of full year in-person visits. We'll see continued acceleration, I would hope. But our business plan calls for having us to educate as many clinicians as possible about the value of our DecisionDx-SCC test in '21. So we enter with a much wider base of ordering doctors than we wouldn't if we just focused on sort of finding the higher volume physicians only and being satisfied with the same number of orders coming in.

I think that's an important distinction in terms of how our company introduces a test, which is clearly going to be transformational in terms of patient care. And that sets us up, I think, for having a wider base of ordering clinicians at the end of '21, since we're in a far nicer position to grow hard in '22, 2024 post-Medicare coverage.

Got it. And then you've talked about the potential to launch these 3 to 5 new tests by the end of 2025. When this year could we learn about what those indications are? And when do you think we could potentially see the first of those tests launch?

First of the test launch, I would say, pick 2025, and then you can rotate around that. It could be that we see a very nice scale-up of the R&D efforts of a couple of our early protocols here in this year and next year such that maybe they -- you might see 1 or 2 launch as early as 2024. But I think safety-wise, '25 at this point in time is a good thing to kind of go around. In terms of timing for sort of discussing specific targets, we're waiting on a couple of data points internally and then thought we would have those conversations here later on in the first part of the year, I believe.

Okay. And then last one from me just on the personalized study. What's the expected enrollment for that study? And when do you think that study could complete and read out?

I think it's a couple of years until we would see a readout. It could come as early as later on -- or in 2022 related this year. But that's probably an aggressive statement, to be honest. So I wouldn't put that in hard cent to hit target-wise. This is essentially a study, which is taking people who can be treated on label with PD-1 inhibitors and assessing the impact of our test to help predict outcomes in those patients. The -- when patients aren't on therapy, the time to recurrence is about, what, 1.2, 1.3 years, I think, from the time of diagnosis for Stage 3 patients through recurrence.

So it shouldn't be that long of a study in terms of reaching an endpoint, but it's too early to kind of make a hard projection there. And part of that, of course, depends upon sort of new patient flow and the sort of COVID time period. We were able to get up and running a number of targeted sites last year. We're quite pleased with that. But to have patients come back for routine blood exams, et cetera, takes a bit of time for that to mature out. I think that will open up quite a bit in the middle of the year going forward.

Our next question comes from the line of Max Masucci from Canaccord Genuity.

First one, just Q1 commentary. Can you just give us a sense, even just directionally, for the step-up you're seeing in DecisionDx-Melanoma reimbursement under the expanded LCD? And are there any factors that might make the modeling exercise for DecisionDx-Melanoma ASPs in 2021 just more complicated than being sort of a black and white process?

Yes, Max. So we -- in the early days here, we are seeing that the expanded LCD criteria covers a little over 90% of the Medicare population. So that's consistent with where we were -- where we expected it to be. So that's good to see. And so that means that of the 45 or so of every 100 patients that are Medicare, we're getting paid on low 40s of those in general.

On the commercial side, it continues to be slow progress. We have -- we still do not have positive policy from the big, large, major -- the 5 big players there, but we are making progress on the regional players. And as we continue our track record of publications and evidence development, we expect to continue to push those over. But as we've said before, it won't be a giant step-wise improvement. It's going to continue to be blocking and tackling.

Okay. Great. And then another just bigger picture one. Just on the AI-based integrated DecisionDx-Melanoma test result. I'm assuming that this is something that your existing customers may have been asking for. And then second part, does the launch of a mobile app sort of suggests that you're positioning the company for success in a post-pandemic world when some of the virtual tools that are in place to help provide continued access to health care might remain in place?

A few questions in there. I think the last one is yes. I think that we've seen increase -- well, even prior to COVID, within dermatology, there are some clinicians who would never pull out their iPhone and forward on something to patients sitting across them in their practice, and some that would. We do see, though, for instance, in the case of dermatologists, there are many clinicians who use an online app to determine if Mohs is an appropriate excision approach for people with basal cell or squamous cell carcinomas. So that's kind of a routine app, which is used clinically with patients. By the way, we see that with other disease states in dermatology.

So part of the sort of feedback that we got even pre-COVID was that when you guys get enough data to provide an integrated report like this, that really takes all the features we think about clinical, pathologic and gene expression profile and using your test and you roll it into a more precise report, we'd love to see that available in terms of an online and a phone-based portal system so that we can make use of that time with our patients when we're seeing them in person.

So I think that's a way going forward, especially with clinicians under 45, under 50 years of age, who've grown up in medical school with that approach in terms of education being part of their life. That's a -- they're working with the trend. We do expect that as we gather data with our other tests that we would continue to roll those into the same platform so that we could provide more and more precise data.

The other earlier part of your comment, I think if I recall, but I'll have you redirect me here when I take a breath, was that we believe there's a certain approach one should take to developing a diagnostic or prognostic test like what Castle does. And the first thing is to see if you can go from using population-based outcome data like staging, clinical and pathologic features. And if the development of a molecular diagnostic test, in our case, a gene expression profile test, actually adds value to what you have in front of you.

So what does that mean is independent of what the doctor and patient already have. But I think that moves you from a population-based to a more personalized-based of health care management, which is fantastic. It's moving you closer and closer to precision-based medicine. I think as we were able to go ahead and successfully generate more prospective consecutively tested patient data in the last year, 1.5 years, 2 years, we were able to validate with an independent sample set the first rollout of this integrated test report in terms of predicting the sentinel lymph node positivity rate of an individual patient based on using all these continuous factors, both the gene expression profile score as well as the clinical pathologic factors.

And that's really moving, I think, from personalized to a more precise medicine approach. And that's the feedback that we have received through market research earlier this year in terms of introducing this new rollout, this new way to think about moving, again, from personalized to much more precision-based medicine. So I think this will be a welcome improvement, our next step, I guess, you would say, in the melanoma test. We have more to do certainly, but it represents a really significant clinical jump forward to really help provide a much more precise single number recommendation with high confidence about what the patient may, may not be facing.

At this time, I am showing no further questions. I would like to turn the call back over to Derek Maetzold, CEO for one final statement.

Thank you, operator. This concludes our fourth quarter and full year 2020 earnings call. Again, I want to thank you for joining us today and for your continued interest in Castle Biosciences.

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.