# Radiation Biology for Nuclear Medicine

## Mechanisms of Radiation Damage

### Direct vs. Indirect Effects

Ionizing radiation damages biological molecules through two distinct mechanisms. In the direct effect, radiation interacts directly with critical biomolecules, particularly DNA, causing ionization of the sugar-phosphate backbone. Direct effects are more common with high-LET radiation such as alpha particles and neutrons. In the indirect effect, radiation first ionizes water molecules, generating highly reactive free radicals including the hydroxyl radical (OH-), hydrogen radical (H-), and hydrated electrons. These free radicals then diffuse a short distance and attack nearby DNA, proteins, and lipids. Indirect effects account for approximately 70% of biological damage from low-LET radiation such as gamma rays, X-rays, and beta particles. The oxygen effect is closely tied to indirect damage: molecular oxygen reacts with radiation-induced free radicals to form organic peroxides that "fix" the damage, making it permanent and more difficult for the cell to repair.

### DNA Damage Types

Radiation produces several types of DNA lesions with varying degrees of biological consequence. Single-strand breaks are the most common lesion and are usually repaired efficiently within minutes to hours. Double-strand breaks are the most critical type of damage because they sever the structural integrity of the DNA molecule and are more difficult to repair accurately. Cells repair double-strand breaks through non-homologous end joining (NHEJ), which is fast but error-prone, or through homologous recombination, which is highly accurate but requires a template strand and is only available during certain cell cycle phases. Other lesion types include base damage (oxidized or deaminated bases repaired by base excision repair), DNA crosslinks (both intrastrand and interstrand), and clustered damage -- the hallmark of ionizing radiation -- where multiple different lesion types occur within one to two helical turns of the DNA. Clustered damage is particularly challenging for cellular repair machinery.

### Linear Energy Transfer (LET)

Linear energy transfer quantifies the density of energy deposition along a radiation track, expressed in keV per micrometer. Low-LET radiation, including gamma rays, X-rays, and beta particles (roughly 0.2 to 2 keV/micrometer), produces sparse ionization and causes damage predominantly through the indirect mechanism. High-LET radiation, such as alpha particles (approximately 100 keV/micrometer) and neutrons, creates dense ionization tracks with more clustered DNA damage that is less amenable to repair. This difference in damage quality makes high-LET radiation more biologically effective per unit of absorbed dose, which is the basis for the higher relative biological effectiveness (RBE) assigned to these radiation types.

## Cell Survival and Dose-Response

### Cell Survival Curves

When cells are irradiated at increasing doses, the surviving fraction follows a characteristic pattern described by the linear-quadratic (LQ) model: S = exp(-alpha*D - beta*D^2). The alpha component represents single-hit lethal damage that increases linearly with dose, while the beta component represents damage from the accumulation of two sublethal events, increasing with the square of the dose. The alpha/beta ratio, the dose at which linear and quadratic contributions are equal, characterizes tissue radiosensitivity. Rapidly dividing tissues such as tumors, bone marrow, and GI epithelium typically have a high alpha/beta ratio of approximately 10 Gy. Late-responding tissues like kidney, spinal cord, and lung parenchyma have a low alpha/beta ratio of approximately 3 Gy, meaning they are more susceptible to damage from larger doses per fraction.

### Factors Modifying Radiosensitivity

Several factors influence how effectively radiation kills cells. The oxygen enhancement ratio (OER) quantifies the increased effectiveness of radiation in the presence of oxygen: for low-LET radiation, oxygen increases damage by a factor of 2.5 to 3, while for high-LET alpha particles, the OER approaches 1, meaning oxygen status is largely irrelevant. This has important implications for therapy: hypoxic tumors are resistant to conventional external beam radiation but remain susceptible to alpha-emitter therapy. Cell cycle phase also matters, with cells in G2/M being most radiosensitive and those in late S phase being most resistant. Dose rate is another modifier: lower dose rates, as encountered in radionuclide therapy compared to external beam radiation, allow cells to repair sublethal damage between hits, generally reducing cell killing. Fractionation of radiation doses similarly allows normal tissue repair between fractions.

## Deterministic Effects (Tissue Reactions)

### Characteristics

Deterministic effects are tissue reactions whose severity increases with dose above a defined threshold. Below this threshold, the effect does not occur because insufficient cells are killed to compromise organ function. These effects can manifest early (hours to months after exposure) or late (months to years), depending on the tissue and dose.

### Examples in Nuclear Medicine

Several deterministic effects are directly relevant to nuclear medicine practice. Bone marrow suppression occurs above a threshold of roughly 0.5 to 1 Gy and manifests as dose-dependent cytopenias. This is a dose-limiting toxicity for many therapeutic radionuclides, including Lu-177 DOTATATE and I-131. Salivary gland dysfunction is a recognized complication of I-131 therapy for thyroid cancer, presenting initially as acute sialoadenitis within 24 to 48 hours and potentially progressing to chronic xerostomia with repeated or high-dose treatments. Radiation nephropathy can result from cumulative kidney doses during peptide receptor radionuclide therapy, with a threshold of approximately 23 to 28 Gy for Lu-177 DOTATATE. Co-infusion of amino acids (lysine and arginine) during therapy reduces renal uptake and provides nephroprotection. Gonadal effects follow a dose-response pattern, with transient oligospermia occurring at 0.15 Gy and permanent sterility at 3.5 to 6 Gy in males, while females experience oocyte loss that is both dose- and age-dependent. Radiation thyroiditis, an acute inflammation of the thyroid gland, typically develops 1 to 2 weeks after I-131 therapy. Pulmonary fibrosis can occur following I-131 treatment in patients with diffuse pulmonary metastases, which is why the whole-body retention at 48 hours is limited to approximately 80 mCi.

### Acute Radiation Syndrome

Acute whole-body exposure above 1 Gy produces the acute radiation syndrome, which progresses through recognizable clinical phases. The hematopoietic syndrome occurs at 1 to 6 Gy and is characterized by pancytopenia. The gastrointestinal syndrome develops at 6 to 10 Gy with denudation of the GI mucosa. The cerebrovascular syndrome occurs above 10 to 20 Gy and causes cerebral edema and death within hours to days. The LD50/60 (dose lethal to 50% of an exposed population within 60 days) is approximately 3.5 to 4.5 Gy without medical treatment and 6 to 7 Gy with supportive care.

## Stochastic Effects

### Characteristics

Stochastic effects differ fundamentally from deterministic effects. The probability of occurrence -- not the severity -- increases with dose. Under the linear no-threshold model, there is no safe threshold dose. Each stochastic effect is all-or-none: either cancer develops or it does not. These effects have a long latency period of years to decades, and a radiation-induced cancer is indistinguishable from a spontaneous cancer.

### Radiation-Induced Carcinogenesis

The latency period for radiation-induced cancer varies by type: approximately 5 to 10 years for leukemia and 10 to 30 or more years for solid tumors. Risk estimates derive primarily from the Life Span Study of Japanese atomic bomb survivors. ICRP Publication 103 provides nominal risk coefficients of approximately 5.5% per Sievert for fatal cancer in the whole population and 4.1% per Sievert for adult workers. The organs with the highest radiation-attributable cancer risk include breast, lung, colon, stomach, and red bone marrow (leukemia). Children are 2 to 3 times more radiosensitive than adults for stochastic effects, owing to their higher mitotic rates and longer remaining lifespan during which a cancer could manifest.

### Hereditary Effects

Radiation-induced germline mutations could theoretically affect future offspring. However, no statistically significant hereditary effects have been documented in any human population, including the extensively studied Hiroshima and Nagasaki survivors. The estimated risk coefficient is approximately 0.2% per Sievert, much lower than the cancer risk. Gonadal shielding and dose optimization remain prudent measures despite the lack of documented human hereditary effects.

## The Linear No-Threshold (LNT) Model

### Principles

The linear no-threshold model assumes that any radiation dose, no matter how small, carries a proportional increase in cancer risk. This model extrapolates the linear relationship observed at high doses in epidemiologic studies down to the low-dose range encountered in diagnostic imaging. It has been adopted by the ICRP, NCRP, and most regulatory bodies worldwide as the basis for radiation protection standards. The model is explicitly conservative, intended for protection purposes rather than precise individual risk estimation.

### Controversy

The LNT model remains scientifically debated. Supporters cite its simplicity, regulatory consistency, the precautionary principle, and molecular evidence that DNA damage occurs at any dose level. Critics point to the adaptive response phenomenon, in which low-dose radiation appears to upregulate cellular DNA repair mechanisms, and to the radiation hormesis hypothesis, which suggests that very low doses might confer a slight net benefit. Epidemiologic studies at low doses (below 100 mSv) lack the statistical power to confirm or refute the LNT model, and the BEIR VII report acknowledged this uncertainty while still endorsing LNT for regulatory purposes. The clinical relevance of this debate is significant: fear of low-dose radiation from diagnostic imaging may lead patients and physicians to inappropriately avoid beneficial studies. Physicians should communicate risk proportionally, noting that diagnostic nuclear medicine doses of 1 to 20 mSv correspond to very small theoretical risks.

## Relative Biological Effectiveness (RBE)

### Definition

Relative biological effectiveness (RBE) is defined as the ratio of the dose of a reference radiation (250 kVp X-rays) to the dose of a test radiation needed to produce the same biological effect. RBE varies with radiation type, the biological endpoint measured, the dose level, and the dose rate. Alpha particles typically have an RBE of 3 to 8 for cell killing, though the radiation weighting factor used for protection calculations is set at 20 to provide a conservative estimate. Beta particles have an RBE of approximately 1.

### Relevance to Radionuclide Therapy

The different RBE values of various radiation types have direct clinical implications for radionuclide therapy. Alpha-emitter therapies using Ra-223, Ac-225, or At-211 achieve high tumor cell killing per unit absorbed dose, are effective against hypoxic tumors (due to their low OER), and have an ideal range of 50 to 100 micrometers for targeting micrometastatic disease. Beta-emitter therapies with Lu-177, Y-90, or I-131 offer a moderate range that provides a crossfire effect, benefiting treatment of tumors with heterogeneous tracer uptake since cells that do not directly bind the radiopharmaceutical can still be irradiated by beta particles from neighboring cells. Auger electron emitters such as I-125 and In-111 have extremely short ranges of less than 1 micrometer and very high LET at the nanometer scale, but they must be internalized into the cell nucleus to achieve maximum therapeutic effect.

| Radiation Type | LET (keV/μm) | Range in Tissue | RBE | OER | Crossfire Effect | Clinical Examples |
|---|---|---|---|---|---|---|
| Alpha particles | ~100 | 50–100 μm | 3–8 | ~1 | Minimal | Ra-223, Ac-225, At-211 |
| Beta particles | 0.2–2 | 1–10 mm | ~1 | 2.5–3 | Significant | Lu-177, Y-90, I-131 |
| Auger electrons | 4–26 | <1 μm | High (if nuclear) | ~1 | None | I-125, In-111 |

<image>A diagram illustrating the direct and indirect mechanisms of radiation-induced DNA damage. Show a DNA double helix with: (1) a gamma ray directly breaking the sugar-phosphate backbone (direct effect), and (2) a gamma ray ionizing a nearby water molecule to produce hydroxyl radicals that then attack the DNA (indirect effect). Include labels for single-strand break, double-strand break, and clustered damage. Show the oxygen enhancement effect where O2 stabilizes free radical damage.</image>

<image>A graph showing cell survival curves for low-LET (X-rays) and high-LET (alpha particles) radiation. The X-ray curve shows a shoulder region (reflecting repair capacity) followed by exponential decline, fitted with the linear-quadratic model. The alpha particle curve shows a steeper, nearly exponential decline with minimal shoulder (less repair). Include the LQ equation and label the alpha and beta components of damage. Show the concept of RBE as the ratio of doses for a given survival level.</image>

<image>A comparative infographic showing the linear no-threshold model alongside alternative dose-response models at low doses. Plot cancer risk on the y-axis versus radiation dose on the x-axis. Show: (1) Linear no-threshold (straight line from zero), (2) Linear-quadratic (curves upward at higher doses), (3) Threshold model (no risk below a threshold), and (4) Hormesis model (slightly decreased risk at very low doses before increasing). Mark the region below 100 mSv where epidemiologic data are insufficient to distinguish between models.</image>

## Clinical Pearls

Double-strand breaks are the most critical DNA lesion caused by ionizing radiation. They are the primary mechanism of both radiation-induced cell killing and mutagenesis, and their repair is inherently error-prone.

Alpha emitters such as Ra-223 and Ac-225 produce high-LET radiation that causes irreparable clustered DNA damage even in hypoxic cells. This explains why alpha-emitter therapy is so potent against micrometastatic disease, where individual tumor cells or small clusters are the targets.

The oxygen effect is clinically important: hypoxic tumors are resistant to low-LET radiation such as conventional external beam or beta-emitter therapy, but they are not resistant to high-LET alpha particles, which have an OER near 1.

Dose-limiting organ toxicity in radionuclide therapy is a deterministic effect with a defined threshold. The key thresholds to know are approximately 23 to 28 Gy for the kidneys with Lu-177 and approximately 2 Gy for the bone marrow with I-131.

The LNT model is a regulatory framework, not a precise calculator of individual risk. Diagnostic nuclear medicine doses of 1 to 20 mSv correspond to tiny theoretical risk increments and should never be a reason to withhold a clinically indicated study.

Children are more radiosensitive than adults by a factor of 2 to 3, reflecting their higher mitotic rates and longer remaining lifespan for cancer expression. Weight-based dose reduction is therefore essential in pediatric nuclear medicine.

When counseling patients about radiation risk from diagnostic studies, it helps to provide context: a 10 mSv study adds approximately 0.05% to the baseline lifetime cancer risk of about 40%.

## References

- Hall EJ, Giaccia AJ. *Radiobiology for the Radiologist*. 8th ed. Wolters Kluwer; 2019.
- ICRP Publication 103. *The 2007 Recommendations of the International Commission on Radiological Protection*. Ann ICRP. 2007;37(2-4).
- National Research Council. *Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2*. National Academies Press; 2006.
- Sgouros G, et al. Radiopharmaceutical therapy in cancer: clinical advances and challenges. *Nat Rev Drug Discov*. 2020;19(9):589-608.