CSTL · Castle Biosciences, Inc.

Current Thesis

Castle Biosciences’ recovery rests on TissueCypher sustaining its growth plan; the next quarterly results must preserve the revenue and test-volume guidance issued on 2026-07-30. At Baird on 2026-09-15, management said its 2026 outlook was unchanged and repeated the 50–52% TissueCypher volume-growth expectation. That addresses the prior concern about the conference passing without substantive confirmation, although it supplies no new quarterly results. Baird conference transcript

The market’s 2026-09-18 reference close of $35.34 exceeds both the $32.91 close on 2026-09-04 and the $34.20 close on 2026-08-21. The inference is that the narrative is maturing — the 2026-09-15 conference reinforced the July outlook, while the same day’s TissueCypher publication added a decision-modeling study rather than reported commercial results. The recovery thesis remains conditional on preserving the August price structure and the published growth expectations. Company study announcement, 2026-09-15

Bullish and bearish views on Castle Biosciences, Inc.

The model's bull view on Castle Biosciences, Inc. (CSTL), in brief: Volume growth supports the recovery. The 2026-07-30 results reported 14,988 TissueCypher tests in Q2 2026 versus 9,170 a year earlier, with companywide revenue of $103.5 million versus $86.2 million. These are measured commercial results supporting the adoption thesis. Q2… The bear view: Dermatology remains an uneven contributor. Both cases follow in full.

Bull Case

• Volume growth supports the recovery. The 2026-07-30 results reported 14,988 TissueCypher tests in Q2 2026 versus 9,170 a year earlier, with companywide revenue of $103.5 million versus $86.2 million. These are measured commercial results supporting the adoption thesis. Q2 results
• The revenue hurdle is explicit. On 2026-07-30, Castle raised full-year revenue guidance to $365–375 million from $345–355 million and forecast positive adjusted earnings before interest, taxes, depreciation and amortization (EBITDA) for Q3, Q4 and full-year 2026. Preserving these expectations at the next results would support the recovery case. Q2 results
• Clinical evidence gained another publication. Castle’s 2026-09-15 announcement described simulated patient-management decisions with and without TissueCypher. This adds evidence concerning clinical use; the announcement does not establish incremental test orders or revenue. Company announcement

Bear Case

• Dermatology remains an uneven contributor. The 2026-07-30 results showed Q2 DecisionDx-Melanoma volume up 3% year over year, DecisionDx-SCC down 16% and MyPath Melanoma down 9%. The measured growth is concentrated in TissueCypher. Q2 results
• Reimbursement still constrains expansion. At Baird on 2026-09-15, management said AdvanceAD-Tx remained in limited launch pending greater confidence in reimbursement. That makes a broader launch conditional on payment economics. Conference transcript
• The equipment order is expenditure. SciBase announced approximately $0.6 million of additional Castle orders on 2026-09-08 for Nevisense Go devices and electrodes supporting clinical studies. This is Castle-funded development activity, not additional Castle revenue. SciBase announcement

Setup & Price Structure

The adjusted market data through 2026-09-18 show a $35.34 close, a three-month price increase of 66.7%, and a close 17.9% below the $43.04 trailing-year high. The 14-day relative strength index was 60.8. These observations establish a substantial recovery; they do not measure investor crowding. Moving-average distances, short interest and fund-flow measurements are unavailable here, so expanding participation cannot be established.

The 2026-08-14 market close of $29.62 remains the dated reference for the August recovery area. A weekly close below $29 would invalidate that structure. This is a research threshold, not a claim that repeated support tests have established a durable floor.

The Form 4 information reported on 2026-09-01 documented Chief Operating Officer Kristen Oelschlager selling shares under a Rule 10b5-1 plan adopted on 2026-03-17. That is observable insider supply during the recovery; the earlier plan date prevents treating the transaction as evidence of a newly formed September outlook. The available sentiment observations are too sparse to support a retail-crowding claim.

Catalyst Calendar (next 30 days)

• 2026-09-19–2026-10-19 review window: At the 2026-09-15 Baird conference, management expected preliminary Medicare pricing for AdvanceAD-Tx in late September. The exact release day is unconfirmed. A proposed rate or an alternative pricing process would clarify the reimbursement path; neither would establish commercial uptake. Conference transcript

As checked on 2026-09-19, Castle’s investor calendar lists no upcoming events and places the 2026-09-15 Baird presentation among past events. The next quarterly-results date remains unconfirmed; no exact catalyst date is assigned. Company event calendar

What Would Change Our Mind

Loss of the August recovery structure would end the price thesis: a weekly close below $29 would put the market beneath the area referenced by the 2026-08-14 close of $29.62. Independently, management reducing the 2026 revenue outlook below the $365 million lower bound issued on 2026-07-30, or cutting the TissueCypher annual volume-growth expectation below 50%, would break the commercial case.

The next quarterly results would confirm the case if Castle preserves both published guidance ranges and reports positive Q3 adjusted EBITDA before the price invalidation occurs. The 2026-09-15 unchanged-outlook statement supports that forecast, but does not substitute for the results. Baird transcript

Correlation Notes

This is a single-company diagnostics thesis anchored in Castle’s 2026-07-30 test volumes and the reimbursement expectations discussed on 2026-09-15. No matched sector-return series is available to establish correlation or attribute the recovery to a broader diagnostics rally.

SciBase’s 2026-09-08 announcement establishes a commercial research relationship, not a measured relationship between the companies’ share prices. The available observations are too few to support a correlation claim. SciBase announcement