Table of Contents
- Key Points
- Why This Research Matters
- How the Study Was Conducted
- Who Took Part in the Study
- What Patients Actually Received
- What the Researchers Measured
- Key Findings: Neutrophil Counts After Chemotherapy
- Key Findings: Severity Grades of Neutropenia
- Key Findings: Tumor Response, Kidneys, Hemoglobin, and Hospital Stays
- Side Effects of Folinic Acid
- What This Means for Patients
- Limitations: What the Study Could Not Prove
- Strengths of the Study
- Practical Recommendations
- Frequently Asked Questions
- Source Information
Key Points
- In a randomized trial of 50 patients with advanced non-small cell lung cancer, oral folinic acid for three days after each pemetrexed cycle protected the bone marrow.
- Median absolute neutrophil count after the first cycle was 3.79 with folinic acid versus 1.85 in controls, a statistically significant difference.
- Low-grade neutropenia was much less common with folinic acid; tumor response, kidney function, and hemoglobin were unaffected.
- Folinic acid was generally well tolerated; the only reported side effect was nausea in three patients, one of whom stopped treatment.
- Folinic acid prophylaxis is promising but not yet standard practice; larger trials are needed before it becomes routine care.
Why This Research Matters
Pemetrexed remains a central chemotherapy drug for non-small cell lung cancer (NSCLC), the most common type of lung cancer. It is combined with immunotherapy (treatments that help the immune system fight cancer) to form the backbone of "chemoimmunotherapy," which has transformed how advanced lung cancer is treated.
But pemetrexed has a serious downside. It suppresses the bone marrow, a condition called myelosuppression (reduced production of blood cells). This is the side effect that limits how much chemotherapy a patient can safely receive.
Even with standard protective vitamins — vitamin B12 and folic acid — up to 26% of patients still develop severe neutropenia (grade III or IV), meaning their infection-fighting white blood cells fall dangerously low. Neutropenia raises the risk of infection, reduces quality of life, can lead to hospitalization or even death, and is a major reason chemotherapy gets delayed or stopped.
The problem is growing worse. Newer treatment combinations are more toxic. When the targeted drug osimertinib is combined with pemetrexed-based chemotherapy (known as the FLAURA2 regimen), 64% of patients experienced a grade III or higher adverse event, and 48% experienced some form of blood-related toxicity. These side effects mean the regimen is often not offered to frail patients.
Pemetrexed toxicity also carries a substantial economic cost. There is, therefore, an urgent need to make pemetrexed safer.
Folinic acid is the already-activated form of folic acid. Unlike ordinary folic acid, folinic acid can directly counteract the toxic effects of pemetrexed. The pemetrexed drug label already recommends high-dose intravenous folinic acid to treat severe neutropenia, and oral folinic acid is approved to prevent blood toxicity from methotrexate — a related antifolate drug. Both drugs work mainly by blocking the enzyme dihydrofolate reductase (DHFR), which is why researchers believe folinic acid should protect against pemetrexed toxicity too.
Until now, though, nobody had systematically tested folinic acid as a routine preventive (prophylactic) measure with pemetrexed. That is exactly what this trial set out to do.
How the Study Was Conducted
This was a randomized, open-label, double-arm study called the FLEX trial. "Randomized" means patients were assigned to treatment groups by chance, which reduces bias. "Open-label" means both patients and doctors knew which treatment each patient received.
The study took place at two large teaching hospitals in the Netherlands: Amphia Hospital and Albert Schweitzer Hospital. The Medical Ethics Committee (METC) Oost-Nederland in Nijmegen, the Netherlands, approved the protocol. The trial followed Good Clinical Practice standards and the Declaration of Helsinki.
The trial was registered publicly: ClinicalTrials.gov number NCT06010277 and European Clinical Trial Registration System (CTIS) number EU-CT 2022-502345-10-00. The funder had no role in study design, data collection, analysis, interpretation, or writing.
Randomization
Participants were assigned in a 1:1 ratio to either the folinic acid group or the control group. A qualified research professional performed the randomization using a validated, secure web-based system (Castor EDC).
Randomization was stratified — that is, balanced — by two factors: the specific chemotherapy regimen and the use of proton pump inhibitors (PPIs, a common class of acid-reducing medicines). This mattered because PPI use is linked to increased pemetrexed toxicity.
Statistical Planning
The researchers used an existing pharmacokinetic/pharmacodynamic (PK/PD) computer model of pemetrexed to run a Monte Carlo simulation — a method that simulates many possible outcomes. The model assumed 100% inhibition of the toxic effect on bone marrow during folinic acid treatment.
The authors are candid that this was an idealized, best-case assumption that may overestimate the real-world effect. A power analysis using a two-sample t-test then calculated that 50 patients were needed to achieve a statistical power of at least 0.80 with an alpha of 0.05. In plain terms, that means the study had an 80% chance of detecting a real difference if one existed, with a 5% risk of a false positive.
Only patients who reached the primary endpoint were included in the analysis (a per-protocol approach). Analysis used IBM SPSS Statistics 29. Baseline characteristics were summarized using medians and interquartile ranges (IQR — the range containing the middle 50% of values). Neutrophil counts, kidney function, and hemoglobin were compared using the Mann-Whitney U test, while rates of neutropenia grades, treatment response, and dose changes were compared using the Chi-squared test.
Who Took Part in the Study
To be eligible, patients had to be 18 years or older, diagnosed with stage IV NSCLC (cancer that has spread beyond the lungs), and treated with pemetrexed at 500 mg/m² combined with platinum-based induction chemotherapy. They also needed an ECOG performance status of 0, 1, or 2 — a standard scale where lower numbers mean better daily functioning.
People were excluded if they had contraindications or drug interactions with folinic acid, a previous allergic reaction to folinic acid or its ingredients, or anemia caused by vitamin B12 deficiency. All patients gave written informed consent.
The two groups were well balanced at the start. Key baseline details:
- Group sizes: 24 patients received folinic acid; 26 served as controls
- Female participants: 11 of 24 (46%) in the folinic acid group vs. 15 of 26 (58%) in the control group
- Median age: 71 years (range 65–73) vs. 72 years (range 67–74)
- Median weight: 70 kg (range 65–84) vs. 68 kg (range 63–77)
- Median height: 171 cm vs. 167 cm
- All patients had NSCLC, and all received pemetrexed at 500 mg/m²
- Median homocysteine (an amino acid used as a marker of folate status): 12.3 μmol/L vs. 14.7 μmol/L
- Median baseline neutrophil count: 5.7 x10⁹/L vs. 6.1 x10⁹/L
- Median estimated glomerular filtration rate (eGFR, a kidney function measure): 89 mL/min vs. 85 mL/min (both capped at 90 in this reporting)
Chemotherapy regimens were distributed evenly. In the folinic acid group, 13 patients (54%) received pemetrexed plus carboplatin plus pembrolizumab; 8 (33%) received pemetrexed plus carboplatin; 3 (13%) received pemetrexed plus cisplatin. In the control group, 14 (54%) received pemetrexed plus carboplatin plus pembrolizumab; 9 (35%) received pemetrexed plus carboplatin; 1 (4%) received pemetrexed plus cisplatin plus pembrolizumab; and 2 (8%) received pemetrexed plus cisplatin.
Proton pump inhibitor use was identical between groups: 12 of 24 (50%) in the folinic acid group and 13 of 26 (50%) in the control group.
Importantly, 5 patients in the folinic acid group had elevated homocysteine levels above 16 μmol/L at baseline, compared with 8 in the control group. This indicated a decreased folate state — a known risk factor for pemetrexed toxicity. Reassuringly, the difference was not statistically significant, meaning the results were not skewed by one group starting out at higher risk.
What Patients Actually Received
The folinic acid group received standard care plus oral folinic acid (brand name Rescuvolin). The dose was 45 mg taken four times daily, starting 24 hours after the pemetrexed infusion. Folinic acid continued for 3 days after each chemotherapy cycle. The control group received standard care without folinic acid.
Everyone in both groups received standard toxicity prevention with folic acid, 0.5 mg once daily. The study period covered four cycles of chemotherapy.
The researchers explain the reasoning behind these specific choices:
- Why 45 mg four times daily: This dose mirrors comparable folinic acid doses used to reduce the toxic effects of methotrexate, which is an even more potent DHFR inhibitor than pemetrexed.
- Why a 24-hour delay: Waiting 24 hours after pemetrexed allowed the drug to be efficiently converted inside cancer cells (a process called polyglutamation), preserving the chemotherapy's anti-cancer effect.
- Why 3 days: A PK/PD analysis showed that patients with a creatinine clearance (a kidney filtration measure) of 45 mL/min — the lowest acceptable kidney function under the pemetrexed label — are most vulnerable to toxicity and can retain toxic pemetrexed levels in the blood for up to 84 hours after infusion. The 3-day folinic acid course covers this worst-case scenario.
Blood samples to measure the absolute neutrophil count (ANC), hemoglobin, and kidney function were collected before each cycle. Additional samples were taken between days 8 and 10 after the first and second cycles, which is when blood counts typically reach their lowest point (the "nadir"). Response CT scans were performed after the second and fourth cycles as part of standard care.
What the Researchers Measured
The primary endpoint was the difference in ANC between the two groups 8 to 10 days after the first cycle of chemotherapy. ANC is a surrogate endpoint — an indirect laboratory marker that stands in for a clinical outcome. Its relevance rests on a well-established, quantitative relationship between how deep and how long neutropenia lasts and the risk of infection.
Secondary endpoints included:
- ANC 8 to 10 days after the second chemotherapy cycle
- The grade of neutropenia according to the Common Terminology Criteria for Adverse Events v5.0 (CTCAE), after both the first and second cycles
Exploratory endpoints included treatment response after the second and fourth cycles (categorized as response, partial response, or progression, assessed locally using RECIST 1.1 criteria), kidney function and hemoglobin before each cycle, reported adverse events from folinic acid, hospital admissions, dose reductions or delays, and discontinuation of chemotherapy. All outcomes from both study sites were centrally assessed and analyzed in the same laboratory.
Key Findings: Neutrophil Counts After Chemotherapy
In total, 64 patients were included between February 7, 2023, and June 28, 2024. Fourteen did not reach the primary endpoint and dropped out, leaving 50 patients in the analysis: 24 in the folinic acid group and 26 in the control group.
The primary endpoint was met decisively. The ANC at nadir after the first chemotherapy cycle was significantly higher (p<0.01) in the folinic acid group:
- Folinic acid group: median ANC 3.79 (IQR 2.22–4.93)
- Control group: median ANC 1.85 (IQR 1.43–3.78)
A p-value below 0.01 means there is less than a 1% probability that a difference this large occurred by chance alone. In practical terms, the folinic acid group's white blood cell counts stayed roughly twice as high.
After the first cycle, 8 patients discontinued treatment — 4 in each group. That left ANC data for the second cycle available for 20 patients in the folinic acid group and 22 in the control group.
The benefit held after the second cycle. The ANC remained significantly higher (p=0.02) with folinic acid:
- Folinic acid group: median ANC 2.60 (IQR 2.03–4.41)
- Control group: median ANC 1.76 (IQR 0.87–2.73)
Key Findings: Severity Grades of Neutropenia
Clinicians grade neutropenia by how low the ANC drops. Grade I is defined as an ANC between 1.5 and 2.0 x10⁹/L; grade II is 1.0 to 1.5; grade III is 0.5 to 1.0; and grade IV is below 0.5 x10⁹/L. As a reference, the generally accepted threshold for "no neutropenia" is above 2.0 x10⁹/L.
After the first cycle, low-grade neutropenia was substantially less common with folinic acid:
- Grade I: 4% with folinic acid vs. 27% in controls (p=0.04 — statistically significant)
- Grade II: 0% vs. 15% (p=0.05 — at the borderline of significance)
- Grade III: 4% vs. 8% (p=0.61 — not significant)
- Grade IV: 0% vs. 4% (p=0.34 — not significant)
After the second cycle, the pattern was less clear-cut:
- Grade I: 11% vs. 19% (p=0.49)
- Grade II: 11% vs. 14% (p=0.74)
- Grade III: 5% vs. 29% (p=0.08 — a trend, not statistically significant)
- Grade IV: 0% vs. 0% (no cases in either group)
That grade III result is worth highlighting even though it did not reach statistical significance. Grade III neutropenia occurred in 5% of folinic acid patients compared with 29% of controls — nearly a six-fold difference in the same protective direction seen with the lower grades. The study was powered only for the primary endpoint, so it lacked the statistical strength to confirm this specific finding.
Key Findings: Tumor Response, Kidneys, Hemoglobin, and Hospital Stays
A crucial question was whether folinic acid might protect the bone marrow at the cost of protecting the tumor as well. The evidence suggests it does not.
No significant difference in cancer response on CT scans appeared between the groups after either the second cycle (p=0.84) or the fourth cycle (p=0.52). The authors note that the study was not powered for this endpoint, so firm conclusions cannot be drawn — but there was no signal of harm.
Other results were similarly neutral:
- Hemoglobin decline from the first to the fourth cycle: median 1.8 (IQR 1.4–2.6) with folinic acid vs. 2.0 (IQR 1.2–2.5) in controls (p=0.32)
- Kidney function decline from the first to the fourth cycle: no significant difference (p=0.57)
- Hospital admissions: 23 in the folinic acid group vs. 18 in the control group (p=0.18) — no significant difference
- Dose delays: no significant difference (p=0.65)
- Treatment discontinuation: no significant difference (p=0.42)
The authors note that the study was underpowered to detect differences in these exploratory secondary endpoints. In other words, the numbers were too small to reliably show an effect even if one existed.
Side Effects of Folinic Acid
Folinic acid was generally well tolerated. The only reported side effects were nausea: two patients experienced grade I nausea (mild) and one patient experienced grade II nausea (moderate), which led that patient to stop taking folinic acid. No other folinic-acid-related adverse effects were reported.
What This Means for Patients
The study's core finding is simple: oral folinic acid taken at home after each pemetrexed infusion kept white blood cell counts significantly higher and reduced low-grade neutropenia. This matters because neutropenia is directly tied to infection risk, hospitalizations, and interrupted cancer treatment.
Because pemetrexed is the backbone of both immunotherapy combinations and EGFR-targeted therapy (such as osimertinib), anything that improves its tolerability could widen access to effective treatment — including for patients who are older or frailer.
Folinic acid may complement, rather than compete with, existing strategies. Granulocyte colony-stimulating factor (G-CSF), for example, is a different type of protection given as an injection. The two work at different points in the same problem:
- Folinic acid acts upstream, directly on the antifolate mechanism. It is taken orally at home, is inexpensive, and is well tolerated.
- G-CSF acts downstream, stimulating white blood cell production. It is used for patients at high risk of febrile or grade III–IV neutropenia, but requires subcutaneous injections, can cause bone pain, and is considerably more expensive.
The authors describe these as complementary approaches targeting different parts of the same clinical problem.
Limitations: What the Study Could Not Prove
The researchers were transparent about what this trial does and does not establish.
- Short follow-up: The study covered only 4 cycles of chemotherapy, roughly 3 months. This limits any conclusions about whether folinic acid protection lasts, or about cumulative bone marrow toxicity during longer pemetrexed maintenance treatment.
- Too short for kidney and hemoglobin effects: The follow-up period was probably too brief to detect meaningful differences in creatinine clearance or hemoglobin, especially since pemetrexed is known to cause kidney toxicity and anemia after longer exposure.
- Open-label design: The primary endpoint was an objective laboratory value (ANC) unaffected by observer expectations. However, the grading and attribution of clinical adverse events were not blinded, nor was the assessment of CT scan responses.
- Underpowered secondary endpoints: Hospitalization, dose delay, and treatment cessation were explored but the study lacked the statistical strength to demonstrate differences.
- Idealized modeling assumption: The sample size calculation assumed 100% inhibition of pemetrexed's bone marrow toxicity. That assumption came from PK/PD modeling and has never been confirmed in human studies. A more conservative estimate would have required a substantially larger group to reliably detect effects on severe neutropenia.
- NSAID use not tracked: Non-steroidal anti-inflammatory drugs (NSAIDs) are linked to increased pemetrexed toxicity and were contraindicated for all participants. Patients were informed of this and reminded by their doctors throughout treatment. However, the absence of formal medication diaries means the researchers could not quantify NSAID exposure as a potential confounding factor. They recommend structured adherence tracking in future studies.
The authors also emphasize that the ANC nadir is a surrogate marker. Translating it into real-world outcomes — fewer infections, fewer hospitalizations, less febrile neutropenia (fever with low white blood cells), and fewer deaths — would require a properly powered trial focused specifically on those endpoints. This proof-of-concept study provides the rationale for that next step.
Strengths of the Study
Several design features strengthen confidence in the results:
- A prospective, multicenter, randomized design
- Stratification by proton pump inhibitor use, which is associated with increased pemetrexed toxicity
- Careful consideration of concomitant (simultaneous) medications
- Measurement of homocysteine levels at baseline — a prognostic biomarker for increased pemetrexed toxicity risk. Because no statistical difference was found between groups, the results are not confounded by one group starting out at higher risk
Practical Recommendations
Based on this trial, patients and their care teams should understand the following:
- Folinic acid prophylaxis is promising but not yet standard practice. This was a small proof-of-concept trial of 50 patients. Larger trials are needed before it becomes routine care.
- Do not self-prescribe folinic acid. The dose used here was specific — 45 mg four times daily, starting 24 hours after pemetrexed, for 3 days. Timing matters, because taking folinic acid too early could theoretically interfere with the chemotherapy's effectiveness.
- Keep taking your standard folic acid and vitamin B12. All participants in this study received folic acid 0.5 mg daily as standard care. Folinic acid was added on top of that, not instead of it.
- Avoid NSAIDs during pemetrexed treatment. These were contraindicated for all study participants because they raise the risk of pemetrexed toxicity. Ask your care team which pain relievers are safe for you.
- Tell your doctor about all medicines you take, including over-the-counter acid reducers (proton pump inhibitors), since these can affect pemetrexed levels.
- Report nausea promptly. Nausea was the only side effect attributed to folinic acid in this study, affecting 3 patients, and it was manageable in most cases.
- Have your blood counts checked on schedule. The critical monitoring window is 8 to 10 days after each infusion, when blood counts typically reach their lowest point.
If you are being treated with pemetrexed and are concerned about low blood counts, ask your oncology team whether a clinical trial of folinic acid prophylaxis is available to you.
Frequently Asked Questions
What did this trial find about folinic acid and pemetrexed chemotherapy?
In a randomized trial of 50 patients with advanced non-small cell lung cancer, taking oral folinic acid for three days after each pemetrexed cycle protected the bone marrow. Median absolute neutrophil count after the first cycle was 3.79 with folinic acid versus 1.85 in controls, a statistically significant difference. Low-grade neutropenia was also less common.
Who was eligible to take part in this trial?
Eligible patients were 18 or older with stage IV non-small cell lung cancer, treated with pemetrexed 500 mg/m² plus platinum-based induction chemotherapy, and an ECOG performance status of 0, 1, or 2. People were excluded for contraindications or drug interactions with folinic acid, previous allergic reaction to it, or anemia from vitamin B12 deficiency.
What exactly did patients in the folinic acid group receive?
They received standard care plus oral folinic acid (Rescuvolin) 45 mg four times daily, starting 24 hours after the pemetrexed infusion and continuing for 3 days after each cycle. Everyone in both groups also received standard folic acid 0.5 mg once daily. The study covered four chemotherapy cycles.
What does an absolute neutrophil count (ANC) of 3.79 mean?
ANC measures infection-fighting white blood cells. In this trial, the median ANC 8 to 10 days after the first cycle was 3.79 in the folinic acid group versus 1.85 in controls. The generally accepted threshold for no neutropenia is above 2.0, so the folinic acid group's counts stayed roughly twice as high.
Did folinic acid reduce the risk of severe neutropenia?
After the first cycle, low-grade neutropenia was much less common with folinic acid: grade I occurred in 4% versus 27% of controls. After the second cycle, grade III neutropenia occurred in 5% of folinic acid patients versus 29% of controls, but this difference was not statistically significant because the trial was powered only for the primary endpoint.
Could folinic acid make the chemotherapy less effective against the cancer?
In this trial, no significant difference in cancer response on CT scans appeared between groups after the second or fourth cycle. The study was not powered for this endpoint, so firm conclusions cannot be drawn, but there was no signal that folinic acid weakened the chemotherapy. Hemoglobin and kidney function were also unaffected.
What side effects did patients experience from folinic acid?
Folinic acid was generally well tolerated. The only reported side effects were nausea: two patients had mild grade I nausea and one had moderate grade II nausea, which led that patient to stop taking folinic acid. No other folinic-acid-related adverse effects were reported in this trial of 50 patients.
If I have advanced non-small cell lung cancer and I'm worried about low white blood cell counts from pemetrexed chemotherapy, when should I seek a second opinion?
A second opinion is reasonable when you have stage IV NSCLC and are starting or continuing pemetrexed, especially if you have had severe neutropenia, dose delays, or treatment stopped because of low blood counts. Folinic acid taken orally for three days after each pemetrexed cycle kept median neutrophil counts higher (3.79 versus 1.85 after cycle one) and reduced low-grade neutropenia, but it is not yet standard practice. A second opinion can clarify whether a clinical trial of folinic acid prophylaxis is available to you. Diagnostic Detectives Network provides independent expert second opinions.
Source Information
Original article title: Prophylactic folinic acid prevents pemetrexed myelosuppression: A randomized trial toward safer treatment with chemotherapy in non-small cell lung cancer.
Authors: Contrucci R, Ter Heine R, Tonn M, Diepstraten J, van Thiel E, van Kesteren C, Heuvel MVD, van der Leest C, de Rouw N.
Affiliations: Department of Pharmacy, Pharmacology & Toxicology, Radboudumc, Research Institute for Medical Innovation, Nijmegen, The Netherlands; Department of Clinical Pharmacy, Amphia Hospital, Breda, The Netherlands; Department of Clinical Pharmacy, Dijklander Hospital, Hoorn, The Netherlands; Department of Pulmonary Diseases, Albert Schweitzer Hospital, Dordrecht, The Netherlands; Department of Clinical Pharmacy, Albert Schweitzer Ziekenhuis, Dordrecht, The Netherlands; Department of Pulmonary Diseases, Radboudumc, Nijmegen, The Netherlands; Department of Pulmonary Diseases, University Medical Center Utrecht, Utrecht, The Netherlands; and Department of Pulmonary Diseases, Amphia Hospital, Breda, The Netherlands.
Publication: Neoplasia, volume 80 (2026), article 101343. Published by Elsevier Inc. under a Creative Commons CC BY license. DOI: https://doi.org/10.1016/j.neo.2026.101343
Dates: Received November 21, 2025; revised July 23, 2026; accepted July 23, 2026.
Trial registration: ClinicalTrials.gov NCT06010277; European Clinical Trial Registration System (CTIS) EU-CT 2022-502345-10-00.
This patient-friendly article is based on peer-reviewed research.
