Exelixis, Inc. (EXEL) - Deep Dive Research Report

Prepared 2026-06-17. Nasdaq: EXEL. Sector: Healthcare / Biopharmaceuticals. Headquarters: Alameda, California (formerly South San Francisco).

Section 1: What the Company Does

Exelixis is a cancer-drug company built almost entirely around one molecule. That molecule is cabozantinib, a tyrosine kinase inhibitor (TKI) - a pill that blocks the enzymes tumors use to grow new blood vessels and to send growth signals. Exelixis discovered it, owns the U.S. commercial rights, sells it under the brand name CABOMETYX, and collects royalties when partners sell it everywhere else in the world. The vast majority of what Exelixis earns comes from this single drug franchise. Everything else the company is doing - a deep clinical pipeline, an aggressive share buyback, a build-out of a second commercial salesforce - is an attempt to answer one question: what happens when cabozantinib's patents expire and its growth runs out?

The company was founded in 1994 by Corey Goodman, Stelios Papadopoulos, and George Scangos. The original idea had nothing to do with cancer pills. Exelixis started as a functional-genomics company, using the genetics of simple model organisms (fruit flies, worms, zebrafish) to systematically map which genes do what, and to find drug targets that way. George Scangos joined as CEO in 1996, set up shop in South San Francisco, and took the company public in 1999. For its first decade Exelixis was a discovery engine that licensed candidate molecules to big pharma partners - it burned cash and had no products of its own.

The pivot that made the company was the decision to keep and develop one of its own compounds, XL184, internally rather than license it out. XL184 became cabozantinib. The first approval came in 2012, for a rare cancer (medullary thyroid cancer) under the brand COMETRIQ. That was a small market. The transformation came when cabozantinib was reformulated as a tablet (CABOMETYX) and approved for advanced renal cell carcinoma - kidney cancer - first in 2016. Kidney cancer is where the franchise became a real business, and CABOMETYX has since become the most-prescribed oral TKI for advanced RCC in the United States.

The core value proposition is straightforward. For an oncologist treating a patient whose kidney cancer has progressed after immunotherapy, or a patient with a neuroendocrine tumor that has nowhere else to go, CABOMETYX is an oral once-daily option backed by survival data, often used alongside an immune checkpoint inhibitor (the Bristol Myers Squibb drug Opdivo). The hard part of this business is not manufacturing a pill - small-molecule synthesis is well understood. The hard part is the two-decade gauntlet of clinical trials, the regulatory approvals, the composition-of-matter and method-of-use patents, and the commercial relationships with oncology practices that you need before that pill generates a dollar. Exelixis spent roughly 18 years and hundreds of millions of dollars getting cabozantinib from a screening hit to a kidney-cancer standard of care.

CEO Michael Morrissey framed the company's ambition on the Q4 2024 call (Feb 11, 2025): the goal is "a multi-compound, multi-franchise oncology business," with aspirational revenue goals of $3 billion for cabozantinib by 2030 and $5 billion for zanzalintinib by 2033.

That second molecule, zanzalintinib, is the whole forward story. It is Exelixis's own next-generation TKI - structurally a successor to cabozantinib, designed with a shorter half-life so it can be combined more safely with immunotherapy. The entire investment case beyond the existing cabozantinib cash flows rests on whether zanzalintinib can become a bigger drug than the one that built the company.

Section 2: Business Segments

Exelixis does not report multiple operating segments. It is a single-segment U.S. oncology pharmaceutical company. Revenue arrives through three economic channels rather than separate divisions, and it is worth separating them because they have very different economics and risk profiles:

1. U.S. net product revenue (the engine). This is CABOMETYX and COMETRIQ sold directly by Exelixis's own commercial organization to U.S. oncology practices, hospitals, and specialty pharmacies. This is the overwhelming majority of the company's revenue and the only channel where Exelixis captures the full price net of rebates. The U.S. cabozantinib franchise grew 17% in 2025, and Q1 2026 set a record for the highest number of new patient starts in any quarter (Q1 2026 call, May 5, 2026). The economics here are very high gross margin - the cost of synthesizing the active ingredient is trivial relative to price - offset by gross-to-net deductions running around 30%, driven by Medicare Part D, co-pay assistance, and a rising share of discounted 340B volume (which crossed 24% of total volume in 2025, per the Q2 2025 call).

2. Collaboration / license revenue (the royalty annuity). Exelixis kept U.S. rights to cabozantinib and licensed the rest of the world. Ipsen holds commercialization rights everywhere outside the U.S. and Japan; Takeda holds Japan. Both pay Exelixis royalties on their sales plus development and regulatory milestone payments. Genentech/Roche partners on COTELLIC. This channel requires no salesforce spend by Exelixis - it is close to pure margin - but Exelixis does not control the pricing or the launch cadence. Ipsen's European Commission approval of CABOMETYX in neuroendocrine tumors (July 2025) is an example of this channel expanding without Exelixis spending commercial dollars.

3. The pipeline (the option value, pre-revenue). Zanzalintinib and the biotherapeutics portfolio generate no revenue today. This is where R&D spend goes and where the company's future franchise value sits. Management treats it as the strategic priority - the reason the company exists beyond cabozantinib's patent cliff.

Because the company is effectively single-product today, there is no segment comparison table to build. The more useful mental model is "one drug paying the bills, one drug being built to replace it, and a royalty stream on the side."

Section 3: Products and Business Detail

CABOMETYX (cabozantinib tablets) - the flagship. A once-daily oral TKI that inhibits multiple receptors (VEGFR, MET, AXL) involved in tumor angiogenesis and progression. U.S. approved indications now span advanced renal cell carcinoma (as monotherapy and in combination with Opdivo), previously treated hepatocellular carcinoma (liver cancer), and - since late March 2025 - previously treated, well-differentiated pancreatic and extra-pancreatic neuroendocrine tumors (NET), based on the CABINET Phase 3 trial. Management has described CABOMETYX as the number-one prescribed TKI in RCC and the market leader in the oral second-line-plus NET segment, with oral TKI TRx share climbing from roughly 40% in early 2024 to about 47% by Q1 2026 (Q1 2026 call, May 5, 2026). The NET launch was notable for speed - the company said it mobilized its salesforce "within hours" of FDA approval and the indication exceeded $100 million in its first partial year (Q4 2025 call, Feb 10, 2026).

COMETRIQ (cabozantinib capsules). The original 2012 formulation, approved for medullary thyroid cancer. A small, mature product - effectively a rounding error against CABOMETYX, but the same molecule.

COTELLIC (cobimetinib). A MEK inhibitor, partnered with Genentech/Roche, used in combination for certain melanoma patients. Exelixis discovered it and earns a share of profits/royalties; it is not a growth driver.

MINNEBRO (esaxerenone). A non-oncology product (a mineralocorticoid receptor blocker for hypertension) that Exelixis out-licensed in Japan, contributing modest royalties. It exists because of the company's earlier discovery output and is not strategically central.

Zanzalintinib (XL092) - the future. A next-generation oral TKI hitting the same target family as cabozantinib (VEGFR, MET, AXL/TAM kinases) but engineered with a shorter half-life, which management argues makes it cleaner to combine with immune checkpoint inhibitors. The development program is wide:

• STELLAR-303 (third-line-plus metastatic colorectal cancer, zanzalintinib + atezolizumab vs regorafenib): met its primary overall-survival endpoint with a 20% reduction in risk of death in the intent-to-treat population; data presented at ESMO 2025 and published in The Lancet; NDA accepted by FDA in February 2026 with a PDUFA target action date of December 3, 2026.
• STELLAR-304 (non-clear-cell RCC, zanza + nivolumab vs sunitinib): fully enrolled, topline results expected 2026, could support a second NDA.
• STELLAR-305 (head and neck cancer): the company discontinued Phase 3 advancement, redeploying resources to indications with what management called roughly threefold greater commercial value and less competition (Q2 2025 call).
• STELLAR-311 (neuroendocrine tumors, zanza monotherapy vs everolimus): Phase 3 enrolling ahead of projections.
• STELLAR-316 (adjuvant/molecular-residual-disease colorectal cancer, using Signatera ctDNA testing): initiation planned around midyear 2026, targeting ~23,000 eligible U.S. patients.
• A Merck collaboration testing zanzalintinib plus belzutifan (Welireg) in RCC.

Biotherapeutics and early pipeline. Exelixis is deliberately broadening beyond small molecules into antibody-drug conjugates (ADCs) and bispecifics: XB010 (a 5T4-targeting ADC), XB628 (a PD-L1/NKG2A bispecific), XB371 (a tissue-factor ADC using site-specific conjugation), XL309 (a USP1 inhibitor for PARP-refractory tumors), and XB064 (an ILT2 antibody). These are Phase 1 or earlier.

Manufacturing and operations. Exelixis relies on contract manufacturers for active pharmaceutical ingredient synthesis and tablet production rather than owning large plants - typical for a small-molecule oncology company. Its differentiated assets are intellectual property (composition and method-of-use patents on cabozantinib, defended through litigation - a favorable IP ruling was highlighted on the Q4 2024 call) and its U.S. oncology commercial infrastructure. In Q1 2026 the company expedited the build-out of a dedicated gastrointestinal (GI) salesforce ahead of the zanzalintinib colorectal launch, separate from its existing genitourinary (GU) cancer salesforce.

Geography. Exelixis itself sells only in the United States. The rest of the world reaches patients through Ipsen (ex-U.S./ex-Japan) and Takeda (Japan), so international expansion shows up as royalty and milestone revenue rather than direct sales.

Section 4: Customers

The economic customer and the decision-maker are different people, which is normal in branded pharma. The prescriber is a U.S. medical oncologist - in academic centers, large community oncology networks, and hospital systems. The payer is the insurer (commercial plans, Medicare Part D, and the 340B discount program for safety-net hospitals). The intermediary is the specialty pharmacy and distributor that physically handles the drug.

The prescribing decision turns on clinical evidence and guideline inclusion. An oncologist reaches for CABOMETYX because it carries Phase 3 overall-survival data, sits in NCCN treatment guidelines, and - critically - has a well-understood combination profile with immunotherapy. In kidney cancer it has become the default oral TKI partner for checkpoint inhibitors. In neuroendocrine tumors it is, per the company, the first and only systemic therapy FDA-approved for previously treated NET regardless of the tumor's site of origin, which is a genuine differentiator because it removes a decision the physician would otherwise have to make.

Switching costs in oncology are not contractual; they are clinical and behavioral. Once an oncologist has a regimen that works and that they are comfortable managing the side effects of, they are slow to change. New entrants must generate their own survival data and earn guideline inclusion, which takes years. That said, this is not a monopoly - oncologists can and do choose lenvatinib (Lenvima), belzutifan (Welireg), or other agents, so the "lock-in" is preference and familiarity rather than a hard moat.

Concentration risk is at the molecule level, not the customer level. No single oncology practice dominates Exelixis's revenue, but a single molecule (cabozantinib) dominates the entire company. The contract structure is essentially spot - prescriptions written month to month - which means revenue is recurring in aggregate (a chronic cancer therapy taken until progression) but has no long-term take-or-pay contracts underpinning it. Predictability comes from the installed base of patients on therapy plus the steady inflow of new patient starts, which is why management reports new-patient-start and TRx/NRx share metrics so prominently.

Section 5: Competitive Landscape

Exelixis competes drug-by-drug, indication-by-indication, against far larger pharmaceutical companies. The structural reality is that Exelixis is a single-product specialist surrounded by diversified giants who can afford to lose a battle in kidney cancer.

In renal cell carcinoma (the core market): the direct rivals to cabozantinib are other oral TKIs and HIF inhibitors used with or without immunotherapy. Lenvatinib (Lenvima), from Eisai partnered with Merck, is the most direct competitor and is frequently paired with Keytruda. Axitinib (Inlyta) from Pfizer is paired with Keytruda. Belzutifan (Welireg), Merck's HIF-2α inhibitor, is a newer mechanism encroaching on later-line RCC. Sunitinib and pazopanib are older generics. Cabozantinib's edge is its MET/AXL activity and its established combination data with Opdivo; its exposure is that Lenvima+Keytruda is a powerful competing regimen and belzutifan represents a mechanistically different threat.

In colorectal cancer (zanzalintinib's lead opportunity): the incumbents in the third-line-plus setting are regorafenib (Stivarga, Bayer) and fruquintinib (Fruzaqla, Takeda). STELLAR-303 was designed head-to-head against regorafenib and won on overall survival, which is the basis for the pending zanzalintinib NDA.

In neuroendocrine tumors: everolimus (Afinitor, now generic), lutetium-based radioligand therapy (Novartis), and somatostatin analogs are the competing approaches; STELLAR-311 tests zanzalintinib directly against everolimus.

Barriers to entry are high but not absolute. They consist of clinical-trial data (years and hundreds of millions of dollars), regulatory approval, patents, and guideline inclusion. They do not protect against a better molecule with better data - which is precisely why Exelixis is racing to build zanzalintinib before cabozantinib's protection erodes. The most important structural shift is the eventual loss of cabozantinib exclusivity and the broader move toward immunotherapy combinations, where the TKI is the junior partner to the checkpoint inhibitor.

Market caps are approximate, for peer-size reference only, as of June 2026, and move with the market.

Section 6: Industry

Exelixis operates in the targeted oncology therapeutics market - specifically the small-molecule and combination-therapy segment for solid tumors. Demand is driven by cancer incidence (which rises with aging populations), by the steady expansion of approved indications as trials read out, and by the shift toward combination regimens that layer a TKI on top of immunotherapy. Unlike a cyclical industrial business, cancer-drug demand is largely non-discretionary and recession-resistant: patients take therapy until disease progression regardless of the economy.

The kidney cancer drug market alone is sizeable and growing. Third-party market researchers put the global kidney cancer drugs market on a path toward roughly $14 billion by the early-to-mid 2030s, growing at high-single-digit to low-double-digit annual rates (Coherent Market Insights; iHealthcareAnalyst). Colorectal cancer - the lead zanzalintinib opportunity - is one of the largest solid-tumor markets by incidence, which is why management frames zanzalintinib as potentially larger than cabozantinib.

The regulatory environment is the defining feature. Every dollar of revenue is gated by FDA approval (and by EMA, PMDA, and other regulators internationally through partners). PDUFA target dates, accelerated-approval pathways, and post-marketing requirements structure the entire calendar. U.S. drug pricing policy is the key headwind: the Inflation Reduction Act's Medicare price-negotiation provisions and the growing 340B discount program both compress net realized prices over time, which is visible in Exelixis's rising gross-to-net deductions. The biggest industry-specific risk for any single-product company is the patent cliff - loss of exclusivity invites generic competition that can collapse a franchise's revenue within a year or two of entry.

Tailwinds at the industry level: an aging population, expanding immunotherapy combinations that pull TKIs along, and ctDNA-guided minimal-residual-disease treatment (the basis for STELLAR-316) opening entirely new earlier-line patient populations. Headwinds: relentless pricing pressure, IRA negotiation, and the high failure rate of late-stage oncology trials.

Section 7: Growth Triggers

All triggers below are drawn directly from the six earnings calls cited. No financial figures are reproduced here.

• Zanzalintinib colorectal cancer approval and launch. FDA accepted the STELLAR-303 NDA (zanza + atezolizumab in previously treated metastatic CRC) with a PDUFA target action date of December 3, 2026. The company expedited the build-out of a dedicated GI salesforce in Q1 2026 to be launch-ready. (Q4 2025 call, Feb 10, 2026; Q1 2026 call, May 5, 2026 - repeated.)

  "In February 2026, the U.S. FDA accepted our NDA for zanzalintinib... with a PDUFA target action date of December 3, 2026." (Q1 2026 call, May 5, 2026)

• STELLAR-304 topline readout (non-clear-cell RCC), expected 2026. Fully enrolled; a positive readout could support a second zanzalintinib NDA. Repeated across multiple calls. (Q1 2025 call, May 13, 2025; Q2 2025 call; Q1 2026 call, May 5, 2026 - repeated.)

• STELLAR-316 initiation (adjuvant/MRD colorectal cancer) around midyear 2026. Uses Signatera ctDNA to select ~23,000 eligible U.S. patients in an earlier-line setting that cabozantinib never reached. (Q1 2026 call, May 5, 2026 - new.)

  Management framed the adjuvant CRC opportunity as roughly a "$1.5 billion market potential." (Q1 2026 call, May 5, 2026)

• STELLAR-311 (NET monotherapy vs everolimus) enrolling ahead of plan. A Phase 3 that could extend zanzalintinib into the neuroendocrine franchise cabozantinib just opened. (Q2 2025 call; Q1 2026 call, May 5, 2026 - repeated.)

• Seven ongoing or soon-to-start zanzalintinib pivotal trials. Management's repeated framing that zanzalintinib could "eclipse the size, scope, and impact" of the cabozantinib business. (Q3 2025 call, Nov 4, 2025; Q1 2026 call, May 5, 2026 - repeated.)

• Merck (belzutifan) and Roche (atezolizumab) combination collaborations. Two RCC studies of zanza + belzutifan launching, plus the atezolizumab CRC combination underpinning the lead filing. (Q4 2024 call, Feb 11, 2025; ongoing - repeated.)

• Neuroendocrine tumor launch ramp (CABOMETYX). Approved late March 2025; salesforce mobilized within hours; indication crossed $100M in its first partial year and continues to ramp; Ipsen secured EU approval (July 2025) extending the indication internationally. (Q1 2025 call, May 13, 2025; Q4 2025 call, Feb 10, 2026 - repeated.)

• Biotherapeutics entering the clinic. XB628 (PD-L1/NKG2A bispecific), XB371 (tissue-factor ADC), XB010 (5T4 ADC), XL309 (USP1) advancing through Phase 1, building optionality beyond TKIs. (Q2 2025 call; ongoing.)

Section 8: Key Risks

Single-product concentration / cabozantinib patent cliff. This is the dominant risk. The overwhelming majority of Exelixis's revenue comes from one molecule. When U.S. composition and method-of-use patents on cabozantinib lapse and generics enter, franchise revenue can fall sharply within one to two years. The entire zanzalintinib program is, in effect, a race against this clock. The mechanism is simple: generic entry collapses price and volume simultaneously. Management's defense - litigating to protect IP (a favorable ruling was flagged on the Q4 2024 call) and building a successor molecule - mitigates but does not eliminate this risk.

Zanzalintinib clinical and regulatory failure. The forward case rests on a molecule that is not yet approved for anything. Late-stage oncology trials fail frequently. A negative STELLAR-304 readout, an FDA complete response letter on the December 3, 2026 PDUFA decision, or disappointing data in the adjuvant CRC or NET studies would directly remove the company's growth narrative. The Q2 2025 discontinuation of STELLAR-305 in head and neck cancer is a live example of the program getting pruned - management framed it as prioritization, but it is also a reminder that not every zanzalintinib indication will work.

Morrissey on the STELLAR-305 decision (Q2 2025 call): the move reflected reallocating to indications with "higher probability of success, little to no competition, and approximately threefold greater commercial value." The candid read: even management's own framing concedes the head-and-neck odds were poor.

Drug pricing and reimbursement pressure. Gross-to-net deductions have climbed to around 30%, driven by Medicare Part D dynamics and a 340B volume mix that exceeded 24% of total in 2025 (Q2 2025 call). The IRA's Medicare negotiation regime is a structural, high-probability, moderate-to-significant drag that compresses net realized price over time regardless of demand growth.

Competitive intensity in RCC. Lenvatinib+Keytruda (Eisai/Merck) and belzutifan (Merck) compete directly for the same kidney cancer patients. A share shift in a major guideline or a strong competitor combination readout could stall the cabozantinib growth that funds the entire enterprise.

Reliance on partners for ex-U.S. value. International revenue depends on Ipsen and Takeda executing launches and pricing; Exelixis does not control those decisions, so a partner's commercial stumble shows up directly in royalty revenue.

Capital allocation execution. The company is spending heavily on R&D and on buybacks simultaneously. If zanzalintinib disappoints, large buybacks executed at higher prices will look like capital that should have funded externally-sourced assets instead - a moderate risk tied to management's own conviction that the stock is undervalued.

Section 9: Walk the Talk

The six calls used for this assessment:

1. Q4/FY2024 - February 11, 2025
2. Q1 2025 - May 13, 2025
3. Q2 2025 - August 2025
4. Q3 2025 - November 4, 2025
5. Q4/FY2025 - February 10, 2026
6. Q1 2026 - May 5, 2026

The most recent (May 5, 2026) is within ~43 days of today, satisfying the recency requirement.

The story across these six calls is one of a management team that has, so far, largely done what it said - on the things it controls. Start with the Q4 2024 call (Feb 11, 2025), where the central promise was launch readiness for the NET indication ahead of an April 3, 2025 PDUFA date. Morrissey said the company was "launch-ready and eager to engage as soon as approval is secured." The drug was approved on March 26, 2025, and by the Q1 2025 call (May 13, 2025) management reported it had mobilized its salesforce within hours and was already taking NET market share. That is a clean kept promise with a verifiable outcome.

On guidance, the team has been credible-to-conservative on cabozantinib. On Q4 2024 the CFO deferred full 2025 net-product-revenue guidance until after NET approval. By Q1 2025 (May 13, 2025) the company raised 2025 net product revenue guidance by $100 million, attributing the bulk to base RCC strength rather than NET, and then held guidance steady through Q2 2025 despite beating - a conservative posture rather than serial guide-raising into hope. The cabozantinib franchise then delivered 17% full-year 2025 growth (Q4 2025 call, Feb 10, 2026), broadly consistent with the trajectory management had been describing.

On the pipeline, the record is more mixed but honest. The big promise repeated from Q4 2024 onward was a STELLAR-303 colorectal cancer readout and, if positive, a fast NDA filing. STELLAR-303 reported positive overall survival on June 22, 2025 (confirmed on the Q2 2025 call), full data landed at ESMO in October 2025 with a Lancet publication (Q3 2025 call, Nov 4, 2025), the NDA was completed before year-end as promised, and the FDA accepted it in February 2026 with a December 3, 2026 PDUFA date (Q4 2025 and Q1 2026 calls). That is a multi-quarter commitment tracked to delivery across four consecutive calls - the strongest evidence of "walking the talk" in the file.

The candor cuts the other way too. On the Q2 2025 call management discontinued STELLAR-305 (head and neck) Phase 3 advancement, having previously flagged it as a 2025 go/no-go decision. They did not bury it - they framed the reallocation explicitly and quantified why (threefold greater value elsewhere). A team that quietly drops a program without acknowledgment is a red flag; Exelixis named the decision and its rationale. Similarly, the STELLAR-304 readout has slipped in framing from "H2 2025" (Q4 2024) to "H1 2026" (Q2 2025) to "2026" (Q1 2026) - an event-driven oncology trial timeline that moved, disclosed each time rather than hidden.

On capital return, management said repeatedly it would buy back stock "when shares are undervalued" and it has executed at scale every single quarter in the window (detailed in Section 10). The CFO's Q1 2026 statement that "we believe Exelixis is undervalued" was backed by ~10 million shares retired that quarter.

Assessment: this is a management team that does what it says on the commercial and capital-return dimensions, and that has been honest about pipeline timelines slipping and programs being cut rather than spinning them. The fair characterization is consistently accurate and conservative on the base business, appropriately candid on the pipeline, with the one open question - the $5 billion zanzalintinib-by-2033 aspiration - still unproven and a long way out.

Section 10: Shareholder Friendliness Index

Dividends. Exelixis pays no dividend and never has. There is no DPS to track over the last three years; the company has chosen to return capital exclusively through buybacks. This is unremarkable for a growth-oriented oncology company that prefers to retain optionality for R&D and business development.

Buybacks and dilution. Exelixis has been an aggressive and consistent repurchaser since launching its first program in March 2023. The history, combining the MoatMap data block (which records zero buybacks in the trailing ~90 days, consistent with no formal program tender in that window but with open-market repurchases continuing) with the company's own filings: the 2023 program retired 26.2 million shares for $550 million by year-end 2023; fiscal 2024 saw roughly $656 million repurchased (with successive $450 million and $500 million authorizations); the board authorized an additional $500 million in February 2025; full-year 2025 repurchases reached $954 million (about 24 million shares at a $39.61 average); and in Q1 2026 the company retired approximately 10 million shares at a $42.99 average, against a $750 million authorization with about $590 million remaining at the end of 2025 (Q4 2025 and Q1 2026 calls; SEC filings). Cumulatively the company has returned roughly $1.9 billion-plus to shareholders via buybacks since March 2023. Critically, this has actually shrunk the share count rather than merely offsetting dilution: weighted-average diluted shares fell from about 326.3 million (2023) to about 278.5 million by Q3 2025 - a reduction of roughly 15%. The MoatMap insider block shows substantial executive equity grants vesting, but the buyback has more than absorbed that dilution.

Verdict: Returns Capital - no dividend, but a sustained, large-scale buyback that has genuinely retired ~15% of the share count over three years while management states the stock is undervalued.

Section 11: Insider Activities

This is a U.S.-listed company (Nasdaq), so the primary source is SEC Form 4 via EDGAR; the MoatMap database block (US venue, current as of 2026-06-17, not flagged stale) is used as the spine and cross-checked for the most recent two weeks. The block covers the trailing 12 months and shows 22 transactions across 14 insiders: zero open-market buys, six open-market sells, and sixteen "Other" (grants/vesting/corporate actions).

Buys - read the signal. There were no open-market purchases by any insider in the last 12 months. There is therefore no bullish conviction-buy signal to flag. This is common for an established, profitable biopharma where executives are already heavily compensated in equity and have little reason to add with cash.

Sells - work out the why. The selling clusters tightly around mid-to-late May and early June 2026, immediately after the May 15 vesting/grant events. The pattern - executives (CFO Senner, R&D head Aftab, Commercial head Haley, GC Hefti) selling within days of equity vesting, and two directors selling in early June - is the classic profile of routine post-vesting liquidation and scheduled disposition rather than a signal about business outlook. The May 27 round of identical 8,367-share, $0-cost grants to nine directors is the annual board equity award following the shareholder meeting, not a market transaction. Specific 10b5-1 plan adoption dates are not reproduced in the MoatMap block; where Form 4 footnotes disclose a plan, these sales are most consistent with scheduled diversification. Reason for the director sells beyond routine portfolio management is not separately disclosed.

Net assessment. Insiders were net sellers over the trailing 12 months, but the activity is broad-based, modest in size relative to each executive's holdings, and tightly correlated with vesting events - the textbook profile of routine compensation-driven selling, not informed bearishness. The most active name was CEO Morrissey, whose transactions were grants and award-related ($0 or vesting), not open-market sales. There is no cluster buying and no first-time CEO purchase to flag. Read: neutral. The absence of any open-market buying means there is no positive conviction signal, but the selling pattern carries no red flag.

Section 12: Scenarios

Bull case. Zanzalintinib works, and it works broadly. The FDA approves it for third-line colorectal cancer on the December 2026 PDUFA date, and the dedicated GI salesforce Exelixis built ahead of time converts the launch into rapid uptake in community oncology. STELLAR-304 reads out positive in non-clear-cell kidney cancer, enabling a second approval, and STELLAR-311 extends the molecule into neuroendocrine tumors. The adjuvant colorectal study (STELLAR-316) opens an entirely new, earlier-line patient population that cabozantinib never touched. In this world, by the late 2020s Exelixis has done exactly what management promised at JPMorgan: a multi-franchise oncology company where zanzalintinib is growing into a bigger product than cabozantinib ever was, the cabozantinib base is still throwing off cash to fund it all, and the relentless buyback has kept the share count shrinking while earnings per share compound. The biotherapeutics portfolio (the ADCs and bispecifics) starts producing its own clinical wins, giving the company a third leg before the second is even mature.

Base case. Cabozantinib keeps doing what it does - growing modestly, holding its leadership in RCC and NET, gradually ceding pricing to the IRA and 340B but offsetting it with new patient starts. Zanzalintinib gets approved in colorectal cancer roughly on schedule and launches respectably, but it takes time to build, and not every STELLAR indication hits - some slip, one or two get pruned the way head-and-neck was. The company remains solidly profitable, the buyback continues to retire shares, and management keeps delivering on the commercial and capital-return promises it has reliably kept. The big question - whether zanzalintinib becomes a $5 billion drug or merely a good one - stays open, and the company looks like a steady, well-run oncology specialist executing a credible succession plan from one franchise to the next, without the explosive re-rating of the bull case.

Bear case. The zanzalintinib bet disappoints. The colorectal approval comes but the commercial uptake is slow because regorafenib and fruquintinib are entrenched and cheap, or the FDA hands down a complete response letter that delays everything. STELLAR-304 or the NET study fails outright, removing the second and third indications that justify the "bigger than cabozantinib" framing. Meanwhile the cabozantinib patent clock keeps ticking: as exclusivity erodes and generics loom, the cash engine that funds the whole pipeline weakens just as the successor stumbles. Lenvima+Keytruda and belzutifan chip away at RCC share, IRA negotiation bites into net price, and the large buybacks executed at $40-50 look, in hindsight, like capital that should have bought an external late-stage asset to fill the gap. In this world Exelixis is a one-drug company watching its one drug mature with no proven replacement - exactly the outcome the entire strategy was built to avoid.

Sources:

• Exelixis Q1 2026 8-K press release (SEC)
• Exelixis Q1 2026 earnings transcript (Motley Fool, May 5, 2026)
• Exelixis Q4/FY2025 earnings transcript (Motley Fool, Feb 10, 2026)
• Exelixis Q4 2025 8-K press release (SEC)
• Exelixis Q3 2025 8-K press release (SEC)
• Exelixis Q2 2025 earnings transcript (Investing.com)
• Exelixis Q1 2025 earnings transcript (Insider Monkey, May 13, 2025)
• Exelixis Q4 2024 earnings transcript (AOL, Feb 11, 2025)
• Exelixis - Our Story (company history)
• Exelixis / Takeda Japan licensing agreement (IR)
• Ipsen expands CABOMETYX collaboration in NET (Ipsen)
• Exelixis FDA accepts zanzalintinib CRC NDA (IR)
• Exelixis beats Bayer in CRC with zanzalintinib (BioSpace)
• Kidney Cancer Drugs Market size (Coherent Market Insights)
• Takeda Fruzaqla (fruquintinib) FDA approval (CancerNetwork)
• Insider transactions: MoatMap multiverse database (US venue, SEC Form 4), current as of 2026-06-17.

A note on scope: I could not write this report to a file (this environment only provides web tools, not file-writing), so the full report and chart-data block are delivered inline above. Section 13 (Further Reading) is intentionally omitted - SemiAnalysis, Stratechery, and MBI Deep Dives have no qualifying coverage of Exelixis, as expected for a biopharma name outside their tech/semis remit.