Can Radiotherapy Affect Routine Biochemical Parameters? Analytical Stability of Serum Control Materials under Therapeutic Photon Exposure

Introduction

Methods

A multi-analyte control solution (Multichem IA Plus; Abbott, Wiesbaden, Germany; lot no. 37109220) was selected, containing 86 analytes spanning multiple diagnostic categories, including biochemical, reproductive and thyroid hormones, steroid hormones, cardiac and anemia markers, therapeutic drug levels, and tumor markers. Although the control material included 86 analytes, only 30 routinely used biochemical parameters that are most commonly applied in daily clinical practice were analyzed in this study.

Since the investigation involved no human or animal-derived samples, ethics committee approval was not required.

Sample Preparation and Irradiation
For each radiation dose, a single aliquot of control solution (n=1) and one non-irradiated control (n=1) were prepared. Radiation doses of 1.5, 1.8, 2, 3.2, 5, 8, 10, 18, and 24 Gy were selected to represent conventional (1.5-2 Gy), hypofractionated (3.2-5 Gy), and stereotactic ablative (>8 Gy) regimens. The non-irradiated aliquot served as the reference control. Equal volumes of each aliquot were sealed to prevent evaporation and contamination, then positioned at the isocenter of a clinical linear accelerator (TrueBeam, Varian Medical Systems, USA).

Prior to irradiation, the dose distribution was verified in the treatment planning system (TPS) using a solid-water phantom model under identical conditions [source-to-surface distance (SSD)=100 cm, 10×10 cm² field] to confirm dose uniformity across the sample region. Since the TPS provides a calculated rather than a directly measured dose, an additional reference ion chamber verification was performed in a solid phantom at the corresponding depth (5 cm water-equivalent). The measured and calculated doses agreed within ±2%, confirming the accuracy of the delivered dose.

Irradiations were delivered using a 6 MV photon beam at a dose rate of 600 cGy/min with a 10×10 cm² field and a SSD of 100 cm. For the horizontal vial geometry (outer diameter ≈ 12 mm), a 1 cm water-equivalent bolus was placed on top of the vial, yielding an effective entrance depth of approximately 1.6 cm to the vial center — near dmax for 6 MV photons. This configuration ensured practical charged-particle equilibrium at the analyte location. The SSD was set to 100 cm at the bolus surface, and the cGy/MU factor was verified under the same geometry.

All aliquots were kept under the same room conditions before and during irradiation, and irradiated samples were analyzed within one hour after exposure. No active temperature monitoring was performed during irradiation; however, because of the small sample volume, sealed vial setup, and short irradiation times, all samples were handled in a standardized manner to minimize preanalytical variability. To minimize potential bilirubin photodegradation, all aliquots were irradiated under dark conditions. Exposure to ambient and direct light was minimized, and all samples were handled using the same irradiation room setup.

Analytical Procedures
The irradiated solutions were analyzed within one hour after irradiation. Each irradiation dose level was represented by a single aliquot of control material, and the 20 replicate measurements reflected technical repeatability of the analytical platform rather than independent biological or experimental replicates. Therefore, the present design was intended to provide a descriptive assessment of immediate analytical stability after irradiation, not a statistical comparison of independently replicated samples. Biochemical assays were performed on the Abbott Alinity c system (photometric, enzymatic, and ion-selective electrode (ISE) methods), while immunoassays were analyzed on the Abbott Alinity i module (chemiluminescent microparticle immunoassay, CMIA). Each analyte was measured using the manufacturer's routine calibrators and internal controls.

Data Processing and Bias Calculation
Each sample was analyzed in 20 technical replicates, and results are presented as mean ± standard deviation (SD). Because each irradiation dose level corresponded to a single aliquot of the same control material measured repeatedly, the dataset did not satisfy the independence assumption required for inferential statistical testing. Consequently, no hypothesis testing or p-values were applied. Instead, analytical performance was assessed descriptively using percentage bias (%Bias) to calculate deviation rates from the target value. A bias of 10% or more was considered clinically significant. %Bias was calculated using the following formula:

Fig 2: Percentage bias (%Bias) calculation formula, where Mean(irradiated) and Mean(control) denote the mean values of the irradiated and control replicates, respectively.

Results

When bias values were assessed across all parameters, 96% remained within ~±6%, indicating strong analytical stability. The only analyte exceeding this range was total bilirubin, which showed a bias of −7.5% at 1.5 Gy and −10.4% at 24 Gy, indicating a gradual dose-dependent decrease (Fig. 1). Measurement precision and reproducibility were preserved across all radiation doses, with no other parameters displaying a dose-related trend.

Fig 1: Variation of total bilirubin (T-Bil) levels with increasing radiation dose.

Liver function enzymes [Aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), and lactate dehydrogenase (LDH); with bias values within ~±4%] remained stable across all dose levels. Although ALT demonstrated a 7.4% reduction after 18 Gy relative to the control, this change remained well within its allowable interference. Enzymatic activity parameters including creatine kinase (CK), amylase, and lipase showed consistent results across all doses, with bias values within ~±3%, indicating that protein conformations and catalytic activity were preserved under irradiation. Creatine Kinase–Myocardial Band (CK-MB) also remained stable, with bias values within −5.8%, confirming no clinically relevant variation.

Renal function markers (urea, creatinine, and uric acid) displayed complete analytical stability up to 24 Gy, with all bias values ~±6%. Lipid profile components [total cholesterol, triglycerides, high-density lipoprotein (HDL) and low-density lipoprotein (LDL)] exhibited only minor fluctuations, with mean bias values within ~±3%, which are clinically insignificant. Electrolytes and minerals (sodium, potassium, chloride, calcium, magnesium, and phosphorus) remained constant at all dose levels, with bias ~±3%, confirming that ISE-based measurements are highly resistant to radiation effects. Finally, total protein, albumin, and C-reactive protein (CRP) concentrations were unaffected by irradiation, maintaining mean bias values within ~±3% relative to the control.

Summary table presenting mean±SD values and bias percentages (%) for all 30 biochemical parameters following exposure to radiation doses between 1.5 and 24 Gy is shown in Appendix 1.

Appendix 1: Summary table presenting mean ± SD values and bias percentages (%) for all 30 biochemical parameters following exposure to radiation doses between 1.5 and 24 Gy

Appendix 1 (Cont.)

Discussion

The preservation of measurement accuracy across diverse analytes—including enzymes, lipids, electrolytes, and acute-phase proteins—suggests that these serum-mimicking control matrices are resistant to radiation-induced degradation. On the other hand, ionized water molecules rapidly undergo proton transfer to produce H₃O+ and •OH, while free electrons simultaneously lose energy and become hydrated electrons—representing the initial stage of radiation-induced chemical reactions.[] Our interpretation is consistent with contemporary radiation chemistry showing ultrafast formation of OH· and e_aq–, while sustained radical propagation typically requires cellular catalysts absent in acellular matrices.[] Also, it has previously been reported that, following gamma sterilization at high doses, various small chemical biomarkers retained more than 90% of their recovery in commercial serum, plasma, and urine samples.[]

Selective bilirubin decline in our study aligns with wavelength-dependent photodegradation kinetics previously described in vitro. The γ-dose-dependent bilirubin breakdown and reduction of the 415 nm heme peak reported by Martel et al.[] are mechanistically consistent with the modest, dose-dependent decrease in total bilirubin observed in our report. The 6 MV photon beam used in clinical RT corresponds to photon energies of approximately 2 MeV (~2×10⁶ eV), nearly one million times greater than those of visible blue light photons (~2.7 eV at 460 nm). Although this large energy difference may appear conceptually confusing, the underlying mechanisms are fundamentally different. While visible light primarily induces non-ionizing photochemical reactions—such as chromophore excitation and photo-oxidation—ionizing X-rays act through direct ionization and the generation of hydroxyl radicals (·OH) and other reactive oxygen species that modify biomolecular structures via oxidative pathways.[] This distinction suggests that analytes containing conjugated chromophore structures are inherently more susceptible to light-induced than radiation-induced degradation.[] Also, in our study, the ~10% change in bilirubin occurred only at the highest dose level (24 Gy). Therefore, it cannot be considered clinically significant. A dose of 24 Gy is commonly used in vertebral SRS treatments; however, a steep dose fall-off occurs around the defined target volume, meaning that direct exposure of body fluids to the full 24 Gy dose is unlikely in most clinical scenarios. Nonetheless, testing such upper-bound doses provides valuable insight into the robustness of analytical stability, extending beyond clinically encountered scatter levels. Moreover, the 7.5% bias observed even at 1.5 Gy suggests that radiation may still exert a subtle but measurable effect. Clinicians should be aware that bilirubin may show mild susceptibility to radiation-associated degradation compared with other routine analytes.

Among the 30 biochemical parameters analyzed, the majority consisted of enzymes and proteins such as AST, ALT, ALP, GGT, CK, LDH, amylase, lipase, and albumin. These macromolecules are composed of amino acid chains and contain catalytically active residues that could theoretically undergo oxidation or denaturation after ionizing radiation exposure.[] However, in our acellular, buffered control matrices, no significant alteration was observed. The absence of intracellular metal ions, thiol-rich environments, and continuous radical regeneration—conditions normally present in living tissues—likely prevented hydroxyl radical propagation and thus protected the tertiary and quaternary structures of these proteins from radiation-induced unfolding.[]

Similarly, non-enzymatic protein derivatives such as urea, creatinine, uric acid, and total protein remained stable after irradiation. These small molecules possess compact chemical structures with no extended conjugated double-bond systems, rendering them relatively resistant to radiolytic fragmentation.[] Lipid-related parameters—including cholesterol, triglycerides, HDL, and LDL—also showed analytical stability. Although lipids are susceptible to peroxidation through radical chain reactions in biological membranes, such processes require sustained oxygenation and catalytic metal ions, which are absent in the serum control solutions used here.[] Similarly, glucose and other carbohydrate components were unaffected, as radiolytic radicals generated in aqueous media rapidly dissipate before inducing glycosidic bond cleavage.[] Electrolytes and inorganic ions (Na+, K+, Cl–, Ca²+, Mg²+, Fe) are elemental species and therefore not subject to chemical decomposition; their measured concentrations remained unchanged across all dose levels.

In clinical settings, patient samples and quality control materials are typically exposed only to very low levels of scattered radiation, generally within the milligray (mGy) to sub-gray (<1 Gy) range. The present study employed a dose spectrum of 1.5-24 Gy, representing the upper bound of potential exposure and designed to assess measurable analytical effects under controlled conditions. Nevertheless, the lowest dose of 1.5 Gy may approximate the theoretical upper limit of scatter doses that could occur near the irradiation field, especially during high-dose or repeated treatment sessions. Even at this near-threshold level, no meaningful analytical deviations were observed, supporting the robustness of most biochemical assays. Previous work has also reported that many routine biochemical analytes remain highly stable under various storage conditions over time.[]

Several limitations should be acknowledged. First, each dose group was represented by a single aliquot, and the 20 replicate measurements corresponded to technical repeatability rather than independent experimental replicates. Accordingly, the present findings should be interpreted as a descriptive assessment of immediate analytical stability rather than a formal statistical comparison of independently irradiated samples. Second, temperature was not continuously monitored during irradiation, although all samples were handled under standardized room conditions and analyzed within the same time window. Third, because bilirubin is known to be photosensitive, irradiation was performed under dark conditions to minimize photodegradation; nevertheless, strict standardization of ambient lighting and LINAC setup conditions remains important for future studies specifically investigating bilirubin degradation. Finally, the use of commercial serum control material, while advantageous for minimizing biological confounding, may limit direct extrapolation to human serum or plasma. Future studies should therefore include multiple independently prepared aliquots, evaluate lower-dose exposures under clinically relevant scatter conditions, and incorporate human-derived samples with extended post-irradiation follow-up to improve translational relevance.

Conclusion

References