Hydrogen therapy can be used to control tumor progression and alleviate the adverse events of medications in patients with advanced non-small cell lung cancer.
Chen JB, Kong XF, Mu F, Lu TY, Lu YY, Xu KC. · Medical Gas Research. 2020;10:285560.
What kind of evidence is this?
Human
Non-randomized comparative clinical study
Cancer research
Inhaled H₂
Mixed observational signal with critical attribution limits.
Clinical symptoms, adverse events and time-to-event outcomes.
Methods at a glance
Patients with advanced non-small cell lung cancer receiving different treatment regimens.
58 enrolled across control, hydrogen-only and hydrogen-plus-drug groups.
Five months.
Inhalation of 66.7% H2 and 33.3% O2 for 4 to 6 hours daily, alone or added to immunotherapy, targeted therapy or chemotherapy.
Mixed H2/O2 inhalation; Brown's-gas-style co-delivery.
Maximum total flow 3,000 mL/min: H2 up to 2,001 mL/min and O2 up to 999 mL/min.
Up to 2,001 mL/min.
Yes — 33.3% O2, up to 999 mL/min.
A small sham-gas control group and non-random treatment cohorts.
Outcomes and reported result
Symptoms, treatment adverse events, tumor response and progression-free survival.
The article reports symptom changes in hydrogen-treated cohorts and tumor or progression outcomes across treatment groups, but comparisons were non-random and groups differed in concurrent therapy.
The full article, group structure, gas specification, follow-up and reported clinical outcomes were checked.
A reported association, difference or mechanism is not automatically a clinical benefit. Null findings, outcome type, study design and precision all matter.
Cautions and applicability
Non-random allocation, small heterogeneous groups, baseline and co-treatment differences and no design capable of isolating hydrogen from drug effects.
The studied advanced-NSCLC cohorts only; not causal evidence of cancer treatment efficacy.
Cochrane RoB 2 — preliminary
Some concerns
Critical confounding by indication and concurrent cancer treatment, with small non-comparable cohorts.
Randomization reported; concealment not fully established · Deviations cannot be excluded · Available-case reporting checked · Outcome measurement varies by endpoint · No outcome-level protocol comparison completed
No single-study GRADE certainty rating is assigned. Certainty is assessed by important outcome across a complete eligible evidence set, not by attaching a final grade to one source.
Sources and record status
PMID: 32541132 · DOI: 10.4103/2045-9912.285560
Free full article in PubMed Central.
PubMed record and PubMed Central full text; extraction checked 9 August 2026.
2026-08-10