Biotech Intelligence Briefing · Monte Rosa · MRT-8102 · MRT-6160 · 2027–2045
Two degraders, no disease efficacy yet — gout has to prove MRT-8102, and MRT-6160 is a milestone stream Novartis controls.
MRT-8102 cuts hsCRP by the same ~80–85% the oral NLRP3 class reports; after ZEUS that carries no CV value on its own, and the first disease-level test is a ~60-patient gout-prevention study reading out in H2 2027. MRT-6160 has seven days of healthy-volunteer dosing behind it; Novartis runs, funds and can end its two Phase 2s, which read out around Dec 2028 and Feb 2029. About 85% of its ~$185M base value to Monte Rosa is milestones. At $11.26 the market prices ~$0.37B for MRT-8102, above our $0.02B base. [dk_design, dk_pd_hscrp, cp_class_hscrp_effect, cp_zeus, cp_clear_synergy, as_gemini1, vv18, vv12, vv13, vv26, vv27, co_px_1001]
Disclosure · The author holds no position in GLUE. Investment research, not investment advice.
Price as of publication · The spot price is the $11.26 close of 1 October 2026, the last close before this briefing was published on 2 October 2026. It is not updated afterwards, so every comparison with spot describes the market on that date.
MRT-8102 human data: Phase 1, 4 weeks, sponsor-reported; PD not placebo-adjusted; no disease-efficacy data. MRT-6160 human data: healthy volunteers, single doses and 7 days; developed by Novartis; licence amounts and rates redacted.
Company Monte Rosa Therapeutics · Nasdaq GLUE Assets MRT-8102 (NEK7 MGD, wholly owned) · MRT-6160 / DDY391 (VAV1 MGD, licensed to Novartis) Status MRT-8102 Phase 1 data reported; Phase 2s proposed · MRT-6160 Novartis Phase 2 (Sjögren's, PsA) Enginemodel/glue.js · engine · loading · HANDOFF-GLUE 2026-10-02 · built from a9d931b (2026-10-05) · inputs 6431298a9ee7 · spot $11.26 (2026-10-01) · facts 315 (16 unverifiable) Data vintage GFORCE-1 2026-10-01 · price 2026-10-01, close $11.26 (Nasdaq) Research cut-off 2026-10-02 (MRT-2359: 2026-10-04) · spot dated 2026-10-01 Author holds no position in GLUE
Scenario
Scenario differences
01
The thesis, in one picture
Risked MRT-8102 revenue by indication for the active scenario, with Monte Rosa's risked MRT-6160 receipts (not Novartis's sales) as a separate line. Dashed verticals are catalysts.
GFORCE-1 showed what a NEK7 degrader does to blood markers in four weeks: hsCRP down a median 85% in treated subjects with week-4 data, within the ~78–86% the oral NLRP3 class reports, with the placebo arm also falling [dk_pd_hscrp, cp_class_hscrp_effect] [unverified: as_placebo_hscrp]. It did not test a disease. The first disease-level test is GEMINI-1, a ~60-patient gout-prevention study with initial data in H2 2027 [as_gemini1]; gout carries the largest base rNPV, and CAD only works if a partner pays for an outcomes trial (§6.2). MRT-6160 is a licence: Novartis funds and runs development, decides Phase 3, owns the lead Sjögren's asset ianalumab and may terminate for convenience [vv12, vv13, vv37]. Monte Rosa's base value from it is mostly milestones [A].
Risked revenue to 2046 and what Monte Rosa receives from MRT-6160
Table: the chart's yearly values
02
Verdict and evidence ladder
One column per asset. No composite score: each rung is a separate question, and "not established" is drawn empty.
Verdict, in plain words
MRT-8102: a potent, low-dose, oral inflammasome-pathway drug with class-level CRP effects and an untested breadth hypothesis. It is not shown to be a better NLRP3 drug, and it is not a CV drug. Strong human target-engagement and PD evidence for an early asset; clinical benefit unproven [dk_nek7, dk_pd_hscrp, cp_class_hscrp_effect].
MRT-6160: strong blood target engagement and ex vivo pathway suppression after seven days in healthy volunteers; no patient data; its value to Monte Rosa is a contract Novartis controls [vv19, vv20, vv18, vv13].
03
GFORCE-1 design and baseline
Phase 1 (Part 3 of the first-in-human study), four weeks of dosing, sponsor-reported. The placebo arm started with lower hsCRP than every active arm.
Design
Randomised 3:1 within each dose cohort, so the n=27 placebo arm pools three cohorts. Not an efficacy trial: the primary endpoint is safety and tolerability.
Baseline hsCRP by arm: median and IQR, mg/L
Placebo 4.8 vs 5.6–6.4 mg/L in active arms. Entry required hsCRP >3 mg/L (dashed line), so regression to the mean is expected in every arm and within-arm % changes are partly regression. Median age 41: a young, obese population, not a CAD population [dk_baseline].
04
Target engagement: NEK7, blood only
A band, not per-arm values: GFORCE-1 arm medians are not disclosed. Peripheral-blood T cells only.
~80–90% NEK7 degradation in peripheral-blood T cells
Medians with IQR in the deck across SAD (n=48), MAD (n=40) and GFORCE-1 week 4 (n=101), at all three GFORCE-1 doses [dk_nek7]. Not measured: monocytes and macrophages, plaque, synovium, HS lesions. Blood T-cell engagement is not tissue engagement. The rabbit gout model reduced swelling at only ~40% degradation [as_preclin_models].
05
Pharmacodynamics at week 4, by biological layer
Median within-arm % change from predose in MRT-8102-treated subjects with week-4 data (n=74 of 81). Not placebo-adjusted. Unknown placebo values are drawn empty, never as zero.
Every marker fell; the hsCRP placebo arm fell too
The analysis set is subjects with week-4 data, n=74 (27 / 26 / 21 by dose); the deck does not call them completers, and the 7 excluded include AE discontinuers, 5 of them at 40 mg, so results may favour tolerators [as_rx_conditions]. Placebo hsCRP read off a figure: ~−16% week 1, −15% week 2, −25% week 3, −28% week 4, ±3 pts [unverified: as_placebo_hscrp]; placebo calprotectin (~−4%) and fibrinogen (~0%) barely moved [as_placebo_other], consistent with regression to the mean on the hsCRP entry criterion.
Table: PD values with set, n, p and source
06
Dose level: a plateau at the lowest dose
No monotonic dose-response on any disclosed marker. AE discontinuations rise with dose on the same axis.
hsCRP and calprotectin by dose, with AE discontinuations
Consistent with saturation at ≤5 mg, matching the SAD/MAD NEK7 plateau [as_barclays_dose]. It does not show that 5 mg is optimal: no lower dose was tested in GFORCE-1 and arm-level NEK7 is undisclosed [dk_nek7]. Bars: median % change (left axis, not placebo-adjusted). Red: AE discontinuations of 27 per arm (right axis) [dk_pd_hscrp, dk_pd_calpro, dk_safety_arms].
The numbers
Interim vs final. The January interim (40 mg only, 24 subjects, active and placebo pooled, split undisclosed) reported −85% and 94% below 2 mg/L [as_interim_40mg]. The final pooled rate is 86% of treated subjects with week-4 data (n=74, three doses) [dk_pd_hscrp]. The two are not comparable. GFORCE-2 doses are 5 / 10 / 15 mg; 20 and 40 mg are dropped [as_gforce2_oct]. The call transcript that may explain why is paywalled and was not read [unverified: as_call_unavail].
07
Safety by arm, 8 weeks of observation
The press release's pooled "33% vs 30%" hides a dose gradient (26 / 30 / 44%) and omits the 8 discontinuations.
View
Adverse events, % of arm (n=27 each)
Counts and discontinuation detail
Four weeks cannot establish chronic safety. Long-term (6-month-study-enabling) toxicology is not complete [as_chronic_tox]. NEK7 cell-cycle biology is an open question: no fact in the base addresses it. Headache 4% vs 11% placebo; arthralgia 5% vs 4% [dk_safety_pooled].
08
Trial-quality audit
What GFORCE-1 can and cannot carry.
Strengths
Limitations
Before GFORCE-1: SAD/MAD in healthy volunteers, preclinical, and the Phase 2 plans
09
Is NEK7 degradation differentiated from direct NLRP3 inhibition?
Duration and placebo handling are always visible. Toggle the treated-arm change or the placebo-handled figure: rows without one are hatched.
Verdict: not shown to be differentiated on today's data
MRT-8102 lowers hsCRP by the same ~80–85% the oral NLRP3 class reports. The only like-for-like duration (28 days, ruvonoflast) shows no measurable gap once placebo is considered, and MRT-8102's own placebo figure is readable only off a chart [dk_pd_hscrp, cp_ruvonoflast_jacc] [unverified: as_placebo_hscrp].
Distinctive but unproven: (a) a dose plateau at 5 mg with clean 5 mg tolerability on 27 subjects; (b) broad DAMP suppression (calprotectin, S100A12, SAA) that comparators have not reported. Neither is a clinical advantage until a disease endpoint moves [dk_safety_arms, dk_pd_calpro, dk_pd_s100a12_saa].
Against colchicine and IL-1/IL-6 biologics the question cannot yet be asked: outcome trials show CRP lowering marks pathway engagement, not benefit, and MRT-8102 has no outcome, imaging or disease data [cp_zeus, cp_clear_synergy].
Defensible claim: a potent, low-dose, oral inflammasome-pathway drug with class-level CRP effects and an untested breadth hypothesis.
Disease yardsticks and deal context
What GEMINI-1 and GALAXY-1 have to beat
NLRP3 deals: context, not inputs
Lilly paid ~$1.2B equity for two Phase 2 NLRP3 assets with biomarker data only [cp_ventyx_lilly_value]. GLUE's EV is ~$0.47–0.52B [co_mcap]. A reference line on the sum-of-parts chart, never a valuation input.
10
What CRP lowering does and does not predict
Forest of MACE hazard ratios with each trial's CRP effect alongside. Different drugs, populations and eras.
11
MRT-6160 and the Novartis licence: the economics
Risked PV to Monte Rosa for the active scenario, driven by the engine. Every redacted amount is hatched with its [A] value in grey, never zero.
US economics (m6_opt_out)
Risked PV of MRT-6160 to Monte Rosa
Table: waterfall components
Licence terms: known vs redacted
The envelope and the structure are hard; every amount, rate, ratio and timing inside the envelope is [A] (§6.3), and the model enforces the envelope in code.
12
Novartis's Phase 2 trials, the timeline and the outcome tree
Registry dates solid; model Phase 3, filing and launch bars hatched and moving with the launch sliders.
Two lanes: Sjögren's and psoriatic arthritis
Joint outcome tree, live
VAV1 biology, dose and safety are shared, so a negative Sjögren's Phase 2 lowers the odds for PsA. Conditional probability in the middle column, joint probability on the right. Phase 3 and approval outcomes are independent across indications given Phase 2 [A simplification]. BIO autoimmune LOA from Phase II is 19.3% [vv47].
Phase 1 card · 7 days, healthy volunteers
Strong blood engagement and ex vivo suppression; nothing about disease efficacy or chronic T/B-cell blockade safety. The ex vivo readouts are stimulation assays.
Preclinical (10-K FY2025, sponsor-reported)
13
Sjögren's and PsA landscape, VAV1 rivals, and the MRT-6160 verdict
Novartis's own assets are tagged. Vendor market sizes are a low-reliability cross-check, never an input.
Sjögren's disease
Psoriatic arthritis
Competing VAV1 degraders
14
MRT-6160 failure points
Live where the engine can read a parameter; hollow where an event is awaited. A pending trigger is never drawn green.
15
MRT-2359: a real signal in a small, selected subgroup
Oral GSPT1 molecular glue degrader + enzalutamide in mCRPC. Sponsor-reported, single-arm, data cut 2026-01-30. Research added 2026-10-04.
MRT-2359 is Monte Rosa's oral GSPT1 degrader, given with enzalutamide in metastatic castration-resistant prostate cancer (mCRPC). An investor note in circulation says the market ignores it. The data show a real-looking signal, but in five patients with androgen-receptor ligand-binding-domain (AR-LBD) mutations, identified after the fact, with no control arm and PSA as the main readout [gs04, gs08, gs37]. The base stays at $50M [A] and the base value per share at $8.42; only the bull moves, from $150M to $250M [A]. The niche is about 400–950 US post-Pluvicto patients a year [unverified: pc_31], and that will not move the needle.
The circulating claims, checked
Claims paraphrased from an investor note in circulation; no one is named or quoted. Verdicts rest only on confirmed or corrected facts. "Untestable" means no source could settle the claim.
Response by subgroup, with exact 95% CIs
Selection and confounding
Why five out of five is weaker evidence than it sounds.
Safety
Phase 1/2 combination cohort; dose reductions not disclosed [gs14].
The mCRPC landscape MRT-2359 would join
Different trials, populations, lines of therapy and endpoints. Larger programmes are already in Phase 3 in the same AR-LBD and post-Lu-PSMA settings.
Small first cohorts tend to shrink
Pluvicto net sales by quarter ($M)
Table: Pluvicto quarterly net sales
Market and value
How big the niche is, and how Monte Rosa treats it.
MRT-2359 failure points
Not advice. The MRT-2359 values are [A] assumptions, not a price target. Not medical advice. Cross-trial numbers are context, not a comparison of efficacy. The author holds no position in GLUE.
16
Which MRT-8102 indication is the best risk-adjusted path?
Live rNPV per indication. Exclude an indication to remove it and its costs (the abandonment option).
17
Funnels: from prevalence to peak sales
Company, independent, derived and assumed layers are styled differently. Not-found layers are dashed, with the assumed value in grey. Sales are never derived from total prevalence.
independent (peer-reviewed or government)company estimatederived (arithmetic on layers above)[A] assumptionnot found → [A] value
18
Headline outcomes
Every tile recomputes from the active scenario or your custom parameters in section 23.
Not advice. Investment research for information only; not investment advice or a recommendation to buy or sell any security. Model outputs are not forecasts or price targets. Not medical advice. Clinical data are summarised for valuation context; nothing here is guidance on any treatment. The author holds no position in GLUE.
19
Sum of parts: what a GLUE share is worth
Per share over 102.885M pre-raise shares (basic + pre-funded warrants + RSUs), then the dilution from the raise and option TSM, against the spot line.
Units
Sum of parts
Table: the bridge
Not advice. The per-share value is the output of stated assumptions, most of them [A]; it is not a price target. Sell-side targets and the Lilly–Ventyx price do not validate it. Not medical advice. The author holds no position in GLUE.
20
Failure-point monitor: MRT-8102 and the bridge
Traffic lights on live values, hollow dots for events the model cannot read yet. MRT-6160 items sit in section 14, MRT-2359 items in section 15.
21
Catalysts, Q4 2026 to 2029
Solid: company guidance. Dotted: registry estimate. Dashed italic: model timing [A]. ★ = one of the five open questions that matter most.
22
Which lever moves the share price?
One at a time: each lever swung from its bear to its bull value with everything else at the active case. The six MRT-6160 milestone amounts move together so the $2.1B envelope holds. Not a confidence interval.
View
Table: lever values
Monte Carlo on value per share
Click the button to populate the histogram · P10 / P50 / P90 appear here
Table: histogram bins
23
Adjust the assumptions
Every slider carries its source and a confidence flag: [H] hard datum · [M] management or filing · [A] analyst assumption · [validated] an [A] cross-checked against an independent benchmark. The milestone sliders refuse values that break the $2.1B envelope. State serialises to the URL.
24
Method, self-tests and caveats
25
What this page does not claim
Shown in full.
That MRT-8102 has been validated clinically, modifies the course of any disease, or lowers cardiovascular risk.
That MRT-8102 lowers hsCRP more than direct NLRP3 inhibitors, or that any cross-trial gap shown is real.
That placebo-adjusted effect sizes for GFORCE-1 are known (the placebo figure is read from a chart).
That biomarkers returned to normal (the healthy-volunteer comparator is separate and non-concurrent).
That Lp(a) −24% is placebo-adjusted, dose-specific, or comparable to PCSK9 inhibitors.
That blood T-cell NEK7 degradation implies degradation in plaque, joint or skin.
That 5 mg is the optimal or commercial dose.
That four weeks of safety predicts chronic safety, or that infection parity over 4 weeks rules out an infection risk.
That CRP lowering predicts MACE reduction.
That company market estimates are independent, or that funnel assumptions are forecasts.
That MRT-8102 has issued composition-of-matter protection to 2044.
That sell-side targets or the Lilly–Ventyx price validate the model.
That MRT-6160 has shown efficacy in any disease, or that seven days in healthy volunteers predict chronic safety of T- and B-cell signalling blockade.
That Novartis's licence, upfront or Phase 2 starts are clinical evidence for MRT-6160, or that Novartis will take it to Phase 3 (it may terminate for convenience and owns ianalumab in Sjögren's).
That any MRT-6160 milestone amount (beyond the $150M upfront and the $50M first Phase 2 milestone), royalty rate, cost-share ratio or opt-out window is known: they are redacted, and the model's values are assumptions inside the disclosed envelope.
That Monte Rosa will co-fund Phase 3, or will opt out.
That MRT-6160 peak sales shown are Monte Rosa revenue (they are Novartis sales), or that any analyst peak-sales or rNPV figure for MRT-6160 exists (none found).
That MRT-6160 is the only clinical-stage VAV1 degrader (Haisco HSK56630 is in Phase 1), or that its patent life is known.
That MRT-2359's 5/5 PSA response applies beyond five post hoc AR-LBD-mutant patients, that a "30% ORR" appears in any disclosure, or that its trial population was post-Pluvicto only.
That MODeFIRe-1 is a registrational study, or that FDA has agreed any path for MRT-2359 (none is disclosed).
That MRT-2359 adds meaningful value: the base keeps it at $50M and the bull at $250M is an option on a niche of ~400–950 US patients a year, our unverified arithmetic.
That MRT-2359 outperforms VIR-5500, pasritamig, opevesostat or any other programme shown beside it.
That the scenario values are forecasts, price targets or recommendations.