RMC-4630

RMC-4630 is an oral SHP2 inhibitor from Revolution Medicines, tested mostly in combination with Amgen's KRAS G12C inhibitor sotorasib for non-small cell lung cancer (NSCLC). Sanofi licensed the drug as SAR442720 in 2018 and returned rights to Revolution Medicines in December 2022 [7]. The Phase 1b combination data (objective response rate 27% in pretreated KRAS G12C NSCLC, 50% in KRAS G12C-inhibitor-naïve patients, presented at WCLC 2022) [5] did not clear the efficacy bar Revolution Medicines set against sotorasib monotherapy (~37% ORR in first-line treatment-naïve), and the company pivoted hard to its direct RAS(ON) inhibitor portfolio (RMC-6236/daraxonrasib, RMC-6291/elironrasib, RMC-9805/zoldonrasib) [6]. The CodeBreak 101 trial flagged here is Amgen's umbrella Phase 1b sotorasib combination protocol [1]. As a standalone asset, RMC-4630 is functionally on the shelf. As a class data point, it tells you something important about why upstream-of-RAS strategies have struggled.

Investigational small molecule. Allosteric SHP2 inhibitor, oral. No approvals anywhere. No FDA breakthrough, fast track, or orphan designations. The drug has run through multiple Phase 1/2 trials: NCT03989115 paired it with cobimetinib (a MEK inhibitor) or osimertinib (EGFR inhibitor) in solid tumors [2]; NCT05054725 was the dedicated Phase 2 in KRAS G12C NSCLC patients who had already failed sotorasib, enrolled 46 patients, and is listed as completed [3]; and it appears as one of many combination arms in Amgen's CodeBreak 101 sotorasib master protocol [1]. Sanofi formally terminated the collaboration on December 7, 2022, with global rights reverting to Revolution Medicines by mid-2023 [7]. Revolution publicly pivoted resources to its direct RAS(ON) program [6]. Mature Phase 2 efficacy data from NCT05054725 was never publicly disclosed at a major conference, consistent with deprioritization rather than a positive readout. There is no announced key trial, no expected regulatory submission, and no near-term readout that would change the trajectory. Treat the 'pipeline' status here as legacy more than active.

SHP2 is a phosphatase - an enzyme that removes the 'off switch' from other proteins. It sits upstream of RAS in the signaling chain that tells cells to grow. Growth factor receptors fire, SHP2 amplifies the signal, and RAS turns on. Block SHP2, and you choke off that amplification. The genetic case is solid: kids born with overactive SHP2 mutations get Noonan syndrome and a sharply elevated leukemia risk, proving SHP2 controls growth signaling [4]. The drug logic for KRAS G12C cancer is more specific. Sotorasib only locks the mutant KRAS protein in its 'off' state. When tumors come back, they often do so by activating wild-type RAS through receptor tyrosine kinase signaling - which runs through SHP2. So SHP2 inhibition was supposed to prevent or reverse sotorasib resistance. The mechanism is biologically clean. The problem is that monotherapy SHP2 inhibition has been weak across every program tried - Novartis (TNO155), Relay/Genentech (RLY-1971/GDC-6971), and others - and the combinations have shown modest, often tolerability-limited activity. The mechanistic reason matters: SHP2 sits downstream of multiple receptor tyrosine kinases simultaneously, so tumor cells under SHP2 pressure can bypass it by amplifying parallel RTK signals or by activating RAS through SOS1-mediated nucleotide exchange independently. That parallel-pathway escape is why monotherapy response was weak and why the combination window with KRAS G12C inhibitors stayed narrow even when the rationale held. The biology validates the target. The drugs have not validated the strategy.

The node points to NCT04185883, CodeBreak 101 - Amgen's Phase 1b master protocol stacking sotorasib with various partners across KRAS G12C tumors, target n=610, primary endpoint dose-limiting toxicity [1]. That is a safety/combination-finding trial, not a registrational study. The more informative trial for RMC-4630 specifically was NCT05054725: Phase 2, single-arm, RMC-4630 plus sotorasib in KRAS G12C NSCLC patients who had already received and failed a KRAS G12C inhibitor (post-KRAS-G12Ci). Primary endpoint was overall response rate (ORR, the fraction of patients whose tumors shrank measurably) by RECIST v1.1. It enrolled 46 patients and is listed as completed [3]. Sample size is small, no comparator arm, post-KRAS-G12Ci is a genuinely hard population. The earlier Phase 1 NCT03989115 ran RMC-4630 with cobimetinib (MEK inhibitor) or osimertinib (EGFR inhibitor) in solid tumors, n=113 [2]. The CodeBreak 101 SHP2 arm reported ORR 27% (95% CI 6-61%, n=11) in patients pretreated with a KRAS G12C inhibitor, and 50% (95% CI 12-88%, n=6) in KRAS G12Ci-naïve patients, with data cutoff April 11, 2022, presented at WCLC 2022 [5]. None of these are powered to support an approval on their own. The design problem is structural - RMC-4630 was always positioned as a partnered asset, never as a standalone registrational program.

Our model puts this drug's chance of eventual approval at 12%. That figure starts from the historical approval rate for Phase 2 drugs in this area - about 13% - and then adjusts up or down based on ten facts about the trial and its sponsor. The design choices (non-randomized, open-label, and more secondary endpoints than usual) nudge the number slightly higher, while weak earlier-phase results pull it back down. Most other factors are close to average for this stage, so the final estimate lands near where the base rate started.

Efficacy is the dominant risk. Every SHP2 inhibitor tested in humans - TNO155, RLY-1971/GDC-6971, JAB-3068, BBP-398, RMC-4630 - has shown weak single-agent activity, and combinations with KRAS G12C inhibitors have produced response rates that did not clearly beat sotorasib alone in the populations tested (Phase 1b ORR 27% in pretreated KRAS G12C NSCLC, vs ~28-37% historical sotorasib monotherapy ORR in second-line and first-line settings respectively) [5]. Mechanism-based GI toxicity (diarrhea, nausea) and thrombocytopenia limit how aggressively you can dose, which caps the combination window. Competitive risk is structural: Revolution Medicines' own direct RAS(ON) inhibitors - RMC-6236 (pan-RAS) and RMC-6291 (G12C-selective) - target the same biology more directly and have shown stronger early data, making RMC-4630 redundant in its developer's own pipeline. Commercial risk is severe. The KRAS G12C NSCLC market is already split between sotorasib and adagrasib, with newer KRAS inhibitors and direct RAS(ON) agents advancing fast. A SHP2 add-on would need to show a clear progression-free survival (PFS, time without tumor growth) or overall survival (OS, time alive) benefit over sotorasib monotherapy to justify the added toxicity and cost, and the data so far have not supported that case. Execution risk is the program itself being quietly wound down rather than formally failed.

Mostly noise. The interesting Revolution Medicines asset is no longer RMC-4630 - it is RMC-6236 (daraxonrasib), the pan-RAS(ON) inhibitor in Phase 3 pancreatic cancer with FDA breakthrough therapy designation, and RMC-6291 (elironrasib), the G12C-selective RAS(ON) inhibitor. RMC-4630 is a useful data point for understanding why the SHP2 class disappointed and why the field moved to direct RAS targeting, but as an investable or trackable program it has run its course. The relevant capital-allocation context is Revolution Medicines' own balance sheet, not Amgen's: Revolution ended Q4 2025 with approximately $2.0B in cash and investments plus up to $2.0B in committed capital from a Royalty Pharma arrangement, against projected 2026 OpEx of $1.6-1.7B [8]. That cushion gives the company room to revive RMC-4630 if a combination rationale with the RAS(ON) inhibitors emerged - for example, SHP2 inhibition specifically preventing or reversing resistance to RMC-6236 or RMC-6291 - but management has given no public signal of such an intention. A real signal would be: an unexpected positive readout from a CodeBreak 101 combination arm involving RMC-4630, a revived development plan from Revolution Medicines or a new partner, or evidence that SHP2 inhibition has a defined role in the RAS(ON) combination strategy. None of those are on the calendar. Check back if Revolution Medicines mentions RMC-4630 on an earnings call again, or if a competitor SHP2 inhibitor produces a surprise positive in a defined biomarker subset. Otherwise, allocate attention to the RAS(ON) program.