The Informed is Root to Rx's deep-dive science track. The Plain Talk newsletter covers the same topic in plain language — subscribe at RootToRx.com. This issue's Root Room: Read Here.

The deeper dive: trial methodology, mechanisms, and the science behind the story.

Branch from Issue 007

In Issue 007, The Informed covered the IRIS trial design, pre-specified stopping rules, and the role of the independent Data Safety Monitoring Board. Issue 008 goes one layer deeper: what happens to a drug's story after approval, when the biology starts writing its own chapters.

The T315I mutation is a substitution at amino acid position 315 in the ABL kinase domain: threonine replaced by isoleucine. Threonine has a hydroxyl group that forms a critical hydrogen bond with imatinib during binding. Isoleucine lacks that hydroxyl group. It also adds steric bulk at that position, physically blocking imatinib from settling into the binding site.

Position 315 is called the gatekeeper residue for a literal reason: it sits at the entrance to the hydrophobic pocket that kinase inhibitors must occupy to bind. Mutations at gatekeeper residues are a general resistance mechanism across kinase-targeted oncology, not unique to BCR-ABL. EGFR T790M, which drove the development of osimertinib in non-small cell lung cancer, operates by the same logic: one substitution at a gatekeeper position, and the first-generation drug can no longer bind.

Nilotinib and dasatinib were engineered with greater conformational flexibility to cover most BCR-ABL resistance mutations. Neither covers T315I at clinically achievable doses. Dasatinib binds BCR-ABL in its active conformation (unlike imatinib, which requires the inactive form), giving it some in vitro activity against T315I, but not at doses tolerated by patients in practice.

Ponatinib, a third-generation TKI, was specifically designed to accommodate T315I using a carbon-carbon triple bond to bypass the isoleucine steric bulk. It works, but carries significant cardiovascular toxicity that has limited its use to patients who have exhausted other options.

Asciminib (Scemblix) targets the myristoyl pocket of BCR-ABL, a hydrophobic regulatory site at the far end of the protein from the ATP-binding site that all prior TKIs targeted. BCR-ABL can exist in active and inactive conformational states. Myristoylation of the native ABL protein's N-terminal region normally helps maintain the inactive state. BCR-ABL in CML has lost this regulation because the BCR-ABL fusion disrupts the N-terminal region. Asciminib binds the myristoyl pocket and allosterically stabilizes the inactive conformation, suppressing kinase activity through a mechanistically distinct approach.

The ASCEMBL trial enrolled patients with CML in chronic phase who had received 2 or more prior TKIs. Major molecular response at 24 weeks: 25.5% (asciminib) vs. 13.2% (bosutinib). At 96 weeks, the difference widened further. For T315I specifically: asciminib at 200mg twice daily is approved for CML in chronic or accelerated phase with confirmed T315I after 1 or more prior TKI. This is the first drug approved specifically for T315I without the toxicity profile of ponatinib for most patients in this setting.

RAS proteins are GTPases: enzymes that cycle between a GTP-bound active state and a GDP-bound inactive state. Oncogenic RAS mutations (KRAS G12D, G12V, G12C, G13C, and others) lock the protein in the GTP-bound state, producing continuous proliferative signaling. The "undruggable" designation was not rhetoric. In its active state, RAS lacks a well-defined binding pocket deep enough for small molecules to occupy with sufficient affinity. Its affinity for GTP is picomolar, far too tight for competitive inhibition by any practical compound.

The breakthrough came from targeting RAS in its inactive or intermediate conformational states, and from exploiting mutation-specific structural features. For KRAS G12C: sotorasib and adagrasib bind covalently to the cysteine introduced by the G12C mutation in the GDP-bound state, locking KRAS inactive. This works, but G12C accounts for roughly 1 to 2% of pancreatic cancer cases.

Daraxonrasib (RMC-6236) is a non-covalent inhibitor targeting the switch II pocket of RAS in the inactive state. Active against multiple KRAS mutations including G12D, G12V, and G13C, which collectively account for the large majority of RAS mutations in pancreatic cancer. This broader coverage is what makes it applicable to a much larger proportion of the disease than G12C-specific agents.

The trial enrolled patients with previously treated metastatic pancreatic cancer with RAS-mutated tumors. Primary endpoint: overall survival. Results: median OS of 13.2 months (daraxonrasib) vs. 6.7 months (chemotherapy). Hazard ratio: 0.40, per the peer-reviewed publication in NEJM (Wolpin BM et al., DOI: 10.1056/NEJMoa2605555, published May 31, 2026, simultaneous with the ASCO plenary presentation).

A hazard ratio of 0.40 means patients on daraxonrasib had a 60% lower risk of death at any given time point compared to the control arm. It does not mean all patients doubled their survival. It is a population-level statistical measure. Individual responses vary considerably. The FDA has opened an expanded access program for eligible patients while the NDA review proceeds.

Tier 1 — Peer-reviewed, published:

IRIS long-term follow-up (Hochhaus A et al., NEJM 2017)

ASCEMBL trial (Rea D et al., NEJM 2021)

Daraxonrasib Phase 1/2 data (Wolpin BM et al., NEJM 2026;394(18):1790-1802. DOI: 10.1056/NEJMoa2505783)

RASolute 302 Phase 3 data (Wolpin BM et al., NEJM 2026. DOI: 10.1056/NEJMoa2605555 — published simultaneously with ASCO plenary, May 31, 2026)

Tier 3 — Company press release (original topline announcement, now superseded):

Revolution Medicines, April 13, 2026. The figures (13.2 vs. 6.7 months, HR 0.40) are consistent with the published trial data.

Tier 4 — Individual case, self-reported:

Ben Sasse's 76% tumor reduction, 60 Minutes interview. Consistent with strong individual responders in the trial. Not the trial average. Not peer-reviewed. Not generalizable.

CML 5-year survival improvement from roughly 30% to roughly 90% after imatinib: peer-reviewed, multiple sources, decades of follow-up.

T315I as the most clinically significant resistance mutation in BCR-ABL therapy: well-established, mechanistically characterized.

Asciminib STAMP mechanism and superiority over bosutinib in ASCEMBL: peer-reviewed, FDA-approved October 2021.

Daraxonrasib Phase 3 OS doubling: Tier 1 as of May 31, 2026. Published in NEJM simultaneously with ASCO 2026 plenary presentation. FDA has opened an expanded access program. NDA review pending.

Ben Sasse individual response: n=1, self-reported. Consistent with strong responders. Cannot be extrapolated to population expectations.

Question 4 at Depth: Revolution Medicines has a financial stake in daraxonrasib. That does not make the data wrong. The IRIS trial was Novartis-funded and the data was real. But it means the peer-reviewed publication and independent ASCO verification serve a specific function: they are the check, not the confirmation. Reading the press release is not the same as reading the trial. Apply the same scrutiny you would apply to anything else. When you do, the evidence here is actually strong.

Question 7: Do I want this to be true, or do I want to know what is true? This question cuts in 2 directions. If you distrust pharmaceutical research, you may want daraxonrasib to fail — to confirm that the pipeline is broken and your skepticism was right. If you are moved by the survival data or a patient's story, you may want it to be more certain than it currently is. Both are forms of motivated reasoning. The discipline is the same in each direction: read the tier of evidence, not the headline.

The funding question has been in the background of every issue in this arc. Issue 009 brings it forward. Who pays for oncology research, what conflicts of interest actually reveal, and what the evidence says when you run the funding question through the same Toolkit.

These 3 issues give you the foundation for everything in this arc.

Issue 003b: How to evaluate any health claim using the Skeptic's Toolkit. Start here.

Issue 001: Why medical distrust is rational, and what to do with it anyway.

Issue 004: How a drug gets from a lab to your pharmacy. The pipeline in plain language.

All back issues are at RootToRx.com.

For educational purposes only. Nothing in this newsletter is medical advice. Talk to your doctor before making any health decision.

Instagram: instagram.com/roottorx | Facebook Lab: facebook.com/groups/theroottorxlab | Subscribe: RootToRx.com

228 Park Ave S, #29976, New York, NY 10003, United States

1. Hochhaus A et al. Long-Term Outcomes of Imatinib Treatment for Chronic Myeloid Leukemia. NEJM. 2017;376(10):917-927.

2. Rea D et al. Asciminib in Chronic Myeloid Leukemia after ABL Kinase Inhibitor Failure (ASCEMBL trial). NEJM. 2021;385(8):721-730.

3. FDA. Scemblix (asciminib) Approval. fda.gov. October 2021.

4. Wolpin BM, Park W, Garrido-Laguna I, et al. Daraxonrasib in Previously Treated Advanced RAS-Mutated Pancreatic Cancer (Phase 1/2). NEJM. 2026;394(18):1790-1802. DOI: 10.1056/NEJMoa2505783.

5. Revolution Medicines. RASolute 302 Phase 3 Results (topline). April 13, 2026. ir.revmed.com.

6. NCT05379985. Phase 1/2 study of daraxonrasib (RMC-6236-001). ClinicalTrials.gov.

7. NCT06625320. RASolute 302 Phase 3 trial. ClinicalTrials.gov.

8. Wolpin BM, Wainberg ZA, Hendifar A, et al. Daraxonrasib or Chemotherapy in Previously Treated Metastatic Pancreatic Cancer (RASolute 302 Phase 3). NEJM. 2026. DOI: 10.1056/NEJMoa2605555.