Reviews/Targeted Therapy Prostate Cancer/PTEN loss and androgen receptor inhibitor resistance in metastatic castration-resistant prostate cancer
Resistance · PC-RES-02 · v2
Borderline: within 4 points of the Likely/Uncertain boundary, so a re-grade could land either side.
This claim says that losing the PTEN gene, and switching on the connected PI3K/AKT growth pathway, makes advanced prostate cancer harder to treat with hormone-blocking drugs and tends to go with worse survival. The strongest evidence here is a blood-based study of 231 men that found patients carrying these genetic changes lived for less time, especially when the changes stacked up alongside other high-risk alterations [2]. A separate randomised trial found that tumours having lost PTEN got more benefit when an AKT-blocking drug was added on top of standard hormone therapy, which fits with PTEN loss helping the cancer resist hormone blockade [1]. The panel treated the survival half of the claim as the better-supported part and the hormone-resistance half as supported but more indirect. It held back from calling this settled because the survival signal is correlative, the resistance signal comes from a surrogate measure in a combination trial, and two other studies in the same corpus point to different genes as the main drivers [6], [13]. The cautious wording 'associated with' is what the evidence here can carry.
The corpus contains one directly prognostic primary study and one directly therapeutic one. [2] profiled 231 men across two independent cohorts (Australian n=78, US n=153) and found PTEN loss in 37% and PIK3CA gain in 17%; patients with dual PTEN loss and PIK3CA gain had a hazard ratio of 2.3 (95% CI 1.2 to 4.4), and cumulative PTEN-PI3K-AKT plus AR copy-number burden tracked with stepwise worse median overall survival (33.5 v 17.2 v 9.7 months and 35.5 v 14.3 v 9.2 months, both P < .001). [1], a randomised phase Ib/II study, showed a larger radiographic progression-free survival gain from adding the Akt inhibitor ipatasertib to abiraterone in PTEN-loss tumours, with overall survival only trending. [13] notes PI3K/AKT mutational status differs between hormone-naive and mCRPC samples. Counterweight comes from [6], which randomised 202 men and tied resistance to BRCA2, ATM, TP53 and AR rather than PTEN/PI3K. The contested point: The division is over whether PTEN/PI3K is a specific, independent driver of AR-inhibitor resistance or simply one correlate of a worse-prognosis genotype. [2]'s survival gradient is built from combined PTEN-PI3K-AKT and AR burden rather than PTEN/PI3K alone, and [6]/[13] name other genes as the dominant resistance and prognostic factors.
All 17 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Targeted Therapy Prostate Cancer, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
[1]'s support for the resistance limb rests on a combination therapy's benefit in PTEN-loss tumours, measured by the surrogate endpoint rPFS in a phase II study with only OS 'trends'; greater benefit from adding an Akt inhibitor does not directly demonstrate that PTEN loss causes resistance to AR inhibitors themselves.
[2]'s survival association is correlative and is driven by cumulative CNV burden across PTEN-PI3K-AKT AND AR pathways jointly; the independent prognostic contribution of PTEN/PI3K alone is not isolated, so the magnitude attributable specifically to the claim's pathway is uncertain.
Where it remains uncertain. The one study purpose-built to find AR-inhibitor resistance drivers implicated BRCA2, ATM, TP53 and AR rather than PTEN/PI3K [6], and another attributed worse prognosis to RB1 loss [13].
Aleatory
The dual PTEN-loss plus PIK3CA-gain subset in [2] and the PTEN-loss arm in the phase Ib/II study [1] rest on limited patient numbers, so chance variation in the effect estimates is substantial.
Epistemic
[1] reports radiographic PFS with overall survival only trending, so the durable clinical benefit of acting on PTEN loss is not yet known from this corpus.
Inferential
The corpus shows PTEN/PI3K associates with worse outcomes [2], but inferring that it specifically predicts AR-inhibitor resistance as a treatment-selection biomarker is a stronger step than the data make.
The load-bearing papers from a corpus of 17, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
States the prognostic association of PTEN loss and co-activation of PI3K-AKT and AR signalling in mCRPC
“A larger rPFS prolongation for the combination was demonstrated in PTEN-loss tumors versus those without.”
Panel auditInternal validity: medium · scope of inference: mCRPC patients receiving abiraterone, comparing PTEN-loss vs PTEN-intact tumors for Akt-inhibitor benefit.Key limitations: no numeric effect sizes or p values · small early-phase study · benefit shown for added Akt inhibitor, not a direct resistance comparison
Quantified prognostic association between PTEN-PI3K-AKT aberrations and inferior survival in mCRPC
“Concurrent CNVs in the PTEN-PI3K-AKT and AR pathways portended poor survival, and identifying this high-risk patient subset for dual AR/Akt inhibition may optimize precision treatment with Akt inhibitors in mCRPC.”
Panel auditInternal validity: medium · scope of inference: mCRPC patients profiled by cfDNA; association of PTEN-PI3K-AKT and AR CNVs with survival.Key limitations: observational association · cfDNA-based, not tissue · resistance to AR inhibitors not directly tested as endpoint
Shows PI3K/AKT pathway alteration evolving in the transition to castration resistance
“Among 61 men with matched hormone-naive and mCRPC biopsies, differences were identified in AR, TP53, RB1, and PI3K/AKT mutational status between same-patient samples.”
Panel auditInternal validity: medium · scope of inference: Genomic evolution from diagnosis to castration resistance in men who develop mCRPC.Key limitations: prognostic signal attributed to RB1, not PTEN · PTEN/PI3K reported descriptively
The abstract does not report PTEN or PI3K/AKT as associated with resistance, so it neither supports nor refutes the claim's driver.
Panel auditInternal validity: medium · scope of inference: Genomic drivers of primary resistance to first-line AR-directed therapy in treatment-naive mCRPC.Key limitations: PTEN/PI3K not reported · focus on DDR and TP53/AR alterations
4 of 17 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Two independent primary studies directly support a prognostic and resistance association, but surrogate endpoints, correlative design and competing drivers within the corpus justify a Likely rather than Established verdict.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Independent prognostic weight of PTEN/PI3K. [2] bundles PTEN-PI3K-AKT with AR copy-number burden, leaving the standalone contribution of the claim's pathway unquantified. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A multivariable survival analysis adjusting for AR-pathway and other high-risk alterations (e.g. TP53, RB1) to isolate PTEN/PI3K's independent effect.
Open question: Does PTEN loss or PI3K/AKT activation independently predict inferior response to AR-pathway inhibitors after adjusting for AR, TP53 and RB1 status? Why it is still open: [2]'s survival signal is measured jointly with AR-pathway burden and [6]/[13] credit other genes, so the independent predictive role is not isolated in this corpus.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 26 Oct 2026 · protocol v0.19