The Survival Gap in CRPC Smokers Showed Up on Enzalutamide

A secondary analysis of Japan’s ENABLE trial found poorer prostate cancer-specific survival among smokers with diabetes or hypertension, and only in the enzalutamide arm. Smoking history by itself did not move survival. The paper, led by urologist Takashi Fukagai of Showa Medical University in Tokyo, went online in the International Journal of Clinical Oncology on 19 September 2026.

That split is the part the headline leaves out. Abiraterone is the androgen-receptor pathway drug that has to be taken with prednisolone, a steroid that can worsen blood sugar. The survival gap among metabolic smokers still showed up on enzalutamide, which does not use a steroid.

Smoking Alone Did Not Change Survival

ENABLE was an investigator-led phase 3 trial that randomly assigned men with castration-resistant prostate cancer, before chemotherapy, to enzalutamide 160 mg a day or abiraterone 1000 mg a day plus prednisolone 5 mg twice daily. Fukagai’s team went back to those records and split patients by smoking habit, then by whether they also had diabetes or hypertension.

THE COUNTS BEHIND THE P VALUE

• Treated set: The secondary analysis reports 178 men randomly assigned and treated.
• Smoking data: Records were complete for 161 men, 84 with no smoking history and 77 with one, leaving 17 without a smoking status.
• Smokers with metabolic disease: In the 77-man smoking group, prostate cancer-specific survival was worse when diabetes or hypertension was present (p = 0.0215).
• Adjusted test: A multivariable Cox model still found diabetes or hypertension to be an independent predictor of poorer cancer-specific survival in that smoking group.

Between never-smokers and men with a smoking history, the authors found no clear difference on the trial’s survival endpoints. The metabolic penalty was not a general smoking effect. It sat inside the smoking group, and it used a combined flag that counted diabetes or hypertension as one exposure.

The Metabolic Penalty Appeared on Enzalutamide

When the smoking group was split by drug, the same cancer-specific survival gap appeared in the enzalutamide arm (p = 0.0201) and did not appear in the abiraterone plus prednisolone arm. The public abstract does not report median survival times or hazard ratios for those slices, so the size of the gap is still behind the journal’s paywall.

SH does not appear to influence the efficacy of ENZ or ABI therapy for CRPC; however, among SH+ patients, the presence of DM or HT was associated with poorer survival outcomes, particularly in those treated with ENZ.

Takashi Fukagai and colleagues, International Journal of Clinical Oncology

That sentence is the finding, and it is also the limit. The analysis can point to poorer cancer-specific survival among smokers who also carried diabetes or hypertension. It cannot say diabetes did the harm by itself, and it cannot say enzalutamide caused the deaths.

Abiraterone Comes With Prednisolone, Enzalutamide Does Not

The two drugs hit the androgen axis at different points. Enzalutamide binds the androgen receptor and blocks signaling inside the cell. Abiraterone blocks CYP17A1, an enzyme the adrenal glands need to make androgens, and that blockade also disturbs cortisol, which is why the ENABLE abiraterone arm took prednisolone.

A separate observational study known as EVADE, in men with metastatic castration-resistant disease, found that abiraterone was linked to new or worsening type 2 diabetes more often than enzalutamide, with a hazard ratio of 1.8. The authors pointed at the steroid. A large Veterans Affairs cohort of men starting these drugs for metastatic castration-sensitive disease later made a related point in older patients: extra glucocorticoid exposure may raise non-cancer deaths even when cancer-specific survival looks similar.

If steroids and blood sugar were the whole mechanism, the ENABLE smoking group should have looked worse on abiraterone. It did not. The cancer-specific split sat on enzalutamide. That inversion is why this paper is more than a diabetes warning.

A 2019 laboratory study offers one path that would fit the drug split. Tobacco smoke condensate raised androgen-receptor variant levels in prostate cancer cells, and smoking exposure increased enzalutamide resistance in those assays. In the same paper’s clinic file, smoking lined up with worse progression-free and cancer-specific survival on androgen-receptor-targeting drugs, not on docetaxel. Abiraterone works upstream of the receptor. A smoking-driven change in the receptor would matter more for a direct blocker. That remains a hypothesis. Fukagai’s paper did not measure receptor variants, pack-years, or glucose control.

Seventy-Seven Smokers Carry the Whole Signal

The original ENABLE report, published in 2022 by Kouji Izumi of Kanazawa University and colleagues, analyzed 92 men in each arm after 203 had been enrolled. It found no overall survival difference between arms. Enzalutamide produced more PSA responses and fewer severe side effects. The secondary analysis’s treated figure is 178, six fewer than that 184-man efficacy set, a gap the open abstract does not explain.

ENABLE PRIMARY RESULTS, 2022

Any-cause adverse events hit 65% of men in each arm. At six months, 96.6% of the enzalutamide arm and 96.7% of the abiraterone arm were alive; at 12 months those figures were 89.0% and 90.8%. Hypertension, diabetes, and high blood lipids were the most common other illnesses at baseline, which is why a later smoking-and-metabolic cut was even possible.

THE ENABLE CALENDAR

1. 1 November 2014: The trial is registered as UMIN000015529, a planned head-to-head comparison before chemotherapy with a target of 100 men per group.
2. 20 February 2015: Enrollment opens across Japanese centres, with randomization run from Kanazawa University’s Innovative Clinical Research Center.
3. 31 July 2019: Enrollment closes after 203 men have been registered.
4. 19 May 2022: The primary paper reports similar survival, a higher PSA response on enzalutamide, and fewer severe events in that arm.
5. 19 September 2026: Fukagai’s secondary analysis posts the smoking and metabolic cut.

The Japanese Foundation for Multidisciplinary Treatment of Cancer paid mainly for the randomization software and, the authors wrote, did not shape the design or the writing. The new analysis was retrospective. It was not powered for a smoking-by-drug-by-metabolic interaction. Seventy-seven smokers, then split again by two drugs and by a combined diabetes-or-hypertension flag, is a thin lattice for a treatment rule.

Diabetes Cuts Incidence Then Complicates Survival

Type 2 diabetes is a strange partner for this cancer. In the CHANCES consortium of older European and American cohorts, prevalent diabetes lined up with a lower chance of a prostate cancer diagnosis (HR 0.81, 95% CI 0.77-0.85). A 45-study meta-analysis led by Bansal, covering 8.1 million people and 132,331 prostate cancers, put the relative risk at 0.86 (95% CI 0.80-0.92). Lower testosterone and sex-hormone-binding globulin, plus metformin’s AMPK effects, are the usual explanations offered for that deficit in new cases.

Once the cancer is present, the picture flips in several older pooled analyses. Cai’s 2015 cohort meta-analysis reported higher prostate cancer-specific mortality in men with diabetes (HR 1.26, 95% CI 1.20-1.33). A 2025 meta-analysis of aggressive disease put the diabetes hazard at 1.18 (95% CI 1.07-1.30). The SEARCH cohort of 976 men with nonmetastatic castration-resistant disease, 304 of them (31%) with diabetes, then showed how messy the label is: coded diabetes and cancer-specific death ran in opposite directions (HR 0.67, 95% CI 0.48-0.92), while diabetes spotted only as high HbA1c without a diagnostic code lined up with higher all-cause death (HR 1.41, 95% CI 1.16-1.72).

Fukagai’s combined diabetes-or-hypertension flag cannot tell those stories apart. A man with well-tracked, treated diabetes and a man with high blood pressure and no glucose data sit in the same “with-DM/HT” bin. The SEARCH pattern is a warning about that bin. Detection and control may matter more than the checkbox.

A Cardio-Oncology Path Is Already in Motion

The authors themselves refuse a prescribing change. They say the work does not show that diabetes causes poorer survival, and it does not show that one of these two drugs should be chosen because of diabetes status. Further prospective studies, they write, would need to test whether tighter metabolic care changes outcomes, and what biological path sits under the smoking-plus-metabolic signal.

WHAT WE KNOW

• The smoking null: In 161 ENABLE patients with a recorded habit, smoking alone did not separate survival on enzalutamide or abiraterone.
• The stacked exposure: Among 77 smokers, diabetes or hypertension predicted worse prostate cancer-specific survival (p = 0.0215), including after multivariable adjustment.
• The drug cut: That cancer-specific gap was detected in the enzalutamide arm (p = 0.0201) and was not detected in the abiraterone arm.

WHAT IS UNCONFIRMED

• Diabetes alone: Hypertension was folded into the same exposure, so an independent diabetes effect is not established.
• A treatment rule: The trial was not built to pick a drug for metabolic smokers, and the authors say it should not be used that way.
• The mechanism: Receptor variants, steroid effects, glucose control, and chance in a small slice are all still on the table.

THE LIMITS OF THIS SIGNAL

• Sample: The entire smoking analysis rests on 77 men, then on still smaller drug slices.
• Design: This is a retrospective cut of a completed trial, not a new randomized comparison.
• Exposure: Smoking pack-years, current versus former use, HbA1c, and blood-pressure control were not the variables on the page.
• Endpoint: The new claim is prostate cancer-specific survival in a subgroup, not a rewrite of ENABLE’s overall survival result.

The practical work sitting next to this paper is less glamorous than a drug switch. On 28 September 2026, cardio-oncologist Avirup Guha and a Prostate Cancer Foundation clinician-industry roundtable published a cardiovascular risk triage schema for men on androgen-deprivation therapy and androgen-receptor pathway inhibitors, arguing that oncology, urology, primary care, and cardiology have to share that risk while cancer treatment continues. Blood pressure, glucose, and smoking status already belong on that shared list. Fukagai’s p-values add a reason to keep them there. They do not settle which capsule the oncologist writes first.

Frequently Asked Questions

What Is Castration-Resistant Prostate Cancer?

ENABLE defined it as a total testosterone level under 50 ng/dL plus two consecutive PSA rises a week apart, with the PSA used for that judgment at least 2 ng/mL above the nadir. Men also had to be free of prior intravenous cytotoxic chemotherapy, be at least 20 years old, and have an ECOG performance status of 0 to 2.

How Do Enzalutamide and Abiraterone Differ?

Enzalutamide is a direct androgen-receptor blocker and is taken without a steroid. Abiraterone shuts down CYP17A1, which cuts androgen production but also disturbs cortisol; ENABLE therefore gave 5 mg of prednisolone twice daily with abiraterone to cover that gap and to limit mineralocorticoid excess such as extra hypertension and low potassium. Luteinizing hormone-releasing hormone therapy continued in both arms, and zoledronic acid or denosumab could be added for bone metastases.

Does Type 2 Diabetes Lower Prostate Cancer Incidence?

Several large pooled analyses say yes for new diagnoses, with relative risks in the mid-0.8s, even as some later studies find higher death rates after the cancer is found. In the SEARCH castration-resistant cohort, diabetes that already had a diagnostic code tracked with lower prostate cancer-specific mortality, while high HbA1c without that code tracked with higher death from any cause, a split that points to detection and treatment quality rather than a single diabetes effect.

What Was ENABLE Designed to Test?

The primary endpoint was time to PSA progression under Prostate Cancer Working Group 2 rules, not overall survival. The protocol targeted 100 men per group, used FACT-G version 4 for quality of life, and allowed investigators to cut the starting dose if the full dose looked wrong for that patient. Sequential treatments after PSA progression were permitted in both arms.

Disclaimer: This article is news reporting and analysis of a published secondary trial analysis and related research. It is for information only and is not medical advice, a diagnosis, or a recommendation to start, stop, or switch enzalutamide, abiraterone, steroids, diabetes drugs, or blood-pressure treatment. Men with prostate cancer, diabetes, or hypertension should talk with their oncologist, urologist, and the clinician who manages their metabolic care before changing any therapy. Figures, p-values, and trial counts reflect the papers and abstracts cited here and may be revised if full data, later follow-up, or new studies appear.