Radiotherapy Principles & Radiation Oncology Basics

Radiotherapy is often presented as a list of doses and machines, which conceals the one idea that makes it comprehensible.

Radiation is not selective. It damages tumour cells and normal cells by the same mechanism. What makes treatment possible is that the two populations differ in how well they repair that damage between exposures.

That difference is the entire therapeutic ratio, and it explains why radiotherapy is given in many small fractions rather than one large dose.

A second idea explains most of the toxicity. Tissues that divide rapidly show effects early; tissues that divide slowly show effects late. Mucositis appears in weeks; fibrosis, vascular damage and second malignancy appear in years.

A third idea governs the clinical decision. The dose that can be delivered is limited not by the tumour but by the normal tissue in the beam, which is why every technical advance in the field has been an attempt to spare that tissue rather than to irradiate the tumour harder.

1. How Radiation Kills

Ionising radiation damages DNA in two ways.

Direct action occurs when the ionising event strikes DNA itself. Indirect action occurs when radiation ionises water, generating free radicals, chiefly the hydroxyl radical, which then damage DNA.

Roughly two-thirds of damage from X-rays is indirect, which is why the presence of oxygen matters so much.

The double-strand break is the lethal lesion, because a single-strand break can be repaired using the complementary strand as a template while a double-strand break cannot.

Cells usually die at the next attempt at division rather than immediately, which is called mitotic death, and this is why rapidly dividing tissues manifest damage first.

Cell cycle position also matters. Cells are most radiosensitive in the G2 and M phases, when chromatin is condensed and repair time before division is short, and most resistant in late S phase, when the machinery for homologous recombination repair is most available.

The oxygen effect

Oxygen fixes radiation damage, meaning it makes the chemical change permanent by reacting with the damaged DNA before it can be chemically restored.

Well-oxygenated cells are roughly two to three times more sensitive than hypoxic ones, expressed as the oxygen enhancement ratio.

This is why hypoxic tumour cores are radioresistant, and it is one of the reasons anaemia worsens outcomes in radiotherapy, since haemoglobin determines oxygen delivery.

It is also an argument for fractionation, because between fractions the tumour reoxygenates as sensitive outer cells die and the remainder come closer to capillaries.

2. Why Fractionation Works

The rationale is captured in five words beginning with R, and each is worth understanding rather than listing.

Repair. Normal tissues repair sublethal damage between fractions more efficiently than tumours, so splitting a dose spares them preferentially. This is the principal reason fractionation works.

Reassortment. Cells are most sensitive in the G2 and M phases, and between fractions surviving cells redistribute through the cycle, so subsequent fractions catch cells in sensitive phases.

Reoxygenation. Hypoxic cells become better oxygenated between fractions as the tumour shrinks, and therefore more sensitive.

Repopulation. Both tumour and normal tissue proliferate during a course. Normal tissue repopulation is helpful and permits recovery; accelerated tumour repopulation is harmful and is the reason interrupting or prolonging a course reduces cure rates.

Radiosensitivity. Intrinsic sensitivity differs between tumour types, which is why lymphoma and seminoma respond to modest doses while melanoma and glioblastoma do not.

The practical consequence of repopulation is examinable and clinically important. Gaps in treatment, whether from machine breakdown, holidays or toxicity, allow surviving tumour cells to proliferate, and prolonging overall treatment time reduces tumour control.

3. Acute and Late Effects

The distinction rests entirely on how fast the target tissue divides.

The last row is the most important and the least intuitive. Late effects are driven principally by the size of each fraction rather than by the total dose, which is why hypofractionated regimens using larger fractions carry a greater risk of late toxicity and why conventional fractionation uses small daily doses.

Second malignancy is the late effect with the longest latency, appearing years to decades later, and it is a particular consideration when treating children and young adults who have a long expected survival.

4. Delivering the Dose

External beam radiotherapy delivers radiation from outside the patient, most commonly using a linear accelerator producing megavoltage photons.

Megavoltage energies have a skin-sparing effect, because maximum dose is deposited a short distance below the surface rather than at it, which is why modern treatment does not produce the severe skin reactions of older orthovoltage machines.

Brachytherapy places the source within or adjacent to the tumour, and its defining advantage follows from physics: dose falls off with the inverse square of distance, so a very high dose can be delivered to the tumour with rapid sparing of surrounding tissue.

This is why brachytherapy is central to cervical cancer treatment, where the source can be placed within the uterine cavity and vaginal fornices, and it is a mainstay in Indian oncology practice.

Systemic radionuclide therapy uses a radioactive agent that concentrates in the target, such as radioiodine for thyroid tissue, which exploits physiology rather than geometry. Samarium and strontium for painful bone metastases and lutetium-labelled agents for neuroendocrine tumours work on the same principle of biological targeting.

Technical advances and what they achieve

Three-dimensional conformal radiotherapy shapes the beam to the tumour outline. Intensity-modulated radiotherapy varies the intensity across each beam, permitting concave dose distributions that wrap around a critical structure. Image-guided radiotherapy verifies position before each fraction, allowing smaller margins.

Every one of these advances spares normal tissue rather than intensifying the tumour dose, which returns to the chapter's third principle.

5. Combining Radiotherapy with Other Treatment

Concurrent chemoradiation exploits synergy, since several agents act as radiosensitisers. Cisplatin is the commonest, used in head and neck, cervical and lung cancers. Toxicity is also synergistic, which is why concurrent treatment demands better performance status than sequential treatment.

Neoadjuvant radiotherapy shrinks a tumour before surgery, improving resectability and, in rectal cancer, reducing local recurrence.

Adjuvant radiotherapy treats microscopic residual disease after surgery, and in breast cancer after breast-conserving surgery it is standard rather than optional.

Palliative radiotherapy is a distinct intention with a distinct technique. It uses fewer, larger fractions over a short time, because late effects are less relevant when survival is limited and because attending daily for weeks is itself a burden. A single fraction is often sufficient for painful bone metastases.

Radiotherapy is genuinely emergency treatment in three situations: spinal cord compression, superior vena cava obstruction and uncontrolled bleeding from a tumour. In cord compression the outcome depends heavily on function at the time treatment starts, which is why dexamethasone and referral do not wait for a formal report.

6. Units, Doses and Tolerance

Three units are used for different purposes and are frequently confused.

The gray measures absorbed dose, meaning energy deposited per unit mass, and it is the unit used in radiotherapy prescribing.

The sievert measures equivalent and effective dose, weighting absorbed dose by the type of radiation and by the sensitivity of the tissue exposed. It is the unit used in radiation protection, because it estimates biological risk rather than physical energy.

The becquerel measures radioactivity, meaning disintegrations per second, and describes a source rather than an exposure.

Typical radical doses

Radical external beam treatment for most epithelial cancers delivers in the region of 60 to 70 gray in fractions of about 2 gray, over six to seven weeks.

Tumours differ enormously in the dose required, which reflects intrinsic radiosensitivity. Seminoma and lymphoma respond to considerably lower doses, while melanoma, sarcoma and glioblastoma are relatively resistant.

Organs at risk

Normal tissue tolerance sets the ceiling, and a few thresholds recur.

The spinal cord is the classic dose-limiting structure, because myelopathy is catastrophic and irreversible, and cord tolerance is one of the first constraints checked in any thoracic or head and neck plan.

The lens is exquisitely sensitive, and cataract is a deterministic effect with a low threshold.

Kidney, lung and liver tolerate whole-organ irradiation poorly, which is why partial-volume treatment and careful planning matter so much when they lie in the field.

The single most important planning question is not how much dose the tumour needs but how much the surrounding tissue will accept, which is why plans are evaluated against organ-at-risk constraints as rigorously as against target coverage.

7. Radiation Protection

The framework has three principles.

Justification: the exposure must do more good than harm. Optimisation: doses should be as low as reasonably achievable, the ALARA principle. Dose limitation: statutory limits apply to workers and to the public, though not to patients receiving treatment.

Protection in practice rests on time, distance and shielding, and distance is the most powerful because dose falls with the inverse square, so doubling the distance quarters the exposure.

Effects are divided into two kinds, and the distinction matters.

Deterministic effects have a threshold below which they do not occur, and severity increases with dose. Cataract, skin erythema, sterility and acute radiation syndrome belong here.

Stochastic effects have no threshold, and dose affects the probability of occurrence rather than the severity. Carcinogenesis and heritable effects belong here.

Pregnancy requires particular care, since fetal risk is greatest during organogenesis, and the guiding rule is that a justified diagnostic study should not be withheld while an unjustified one should never have been requested. Therapeutic radiotherapy to the abdomen or pelvis is quite different from a diagnostic exposure and is generally incompatible with continuing a pregnancy, which makes the discussion a shared decision rather than a technical one.

8. Worked Examples

Example 1. A patient receiving radical radiotherapy for head and neck cancer develops severe mucositis at week four and the team considers a two-week break. What is the concern?

Accelerated repopulation. Tumour clonogens proliferate during a course of radiotherapy, and in head and neck squamous carcinoma this acceleration typically begins around three to four weeks into treatment, so surviving cells are dividing considerably faster than they were at the start.

A gap therefore allows the tumour to regain ground that already-delivered fractions had taken, and prolonging overall treatment time is associated with reduced local control and worse survival. Estimates suggest a measurable loss of tumour control for each additional day of overall treatment time.

The alternative is to manage the mucositis aggressively rather than to interrupt: analgesia including opioids, mouth care, nutritional support with a feeding tube where necessary, and treatment of superimposed candidal infection.

If a gap becomes unavoidable, compensation is attempted by accelerating the remainder or adding fractions, and the reasoning is documented.

Example 2. Why does anaemia reduce the effectiveness of radiotherapy?

Because roughly two-thirds of the DNA damage caused by X-rays is indirect, mediated by free radicals generated from water, and oxygen is required to make that damage permanent.

The mechanism is called oxygen fixation. When a hydroxyl radical produces a DNA lesion, that lesion can be chemically restored by hydrogen donation from cellular thiols. If oxygen is present it reacts with the damaged site first, forming an organic peroxide and rendering the damage permanent and therefore lethal.

Well-oxygenated cells are consequently two to three times more sensitive than hypoxic ones, which is the oxygen enhancement ratio.

Anaemia reduces oxygen delivery to the tumour, increasing the hypoxic fraction and effectively making the tumour more radioresistant. Observational data associate low haemoglobin with poorer local control in several tumour sites, particularly cervical and head and neck cancer.

The practical response is to correct treatable causes of anaemia, though trials of aggressive correction with erythropoiesis-stimulating agents have not shown benefit and have raised safety concerns, so transfusion decisions follow standard indications rather than being driven by the radiotherapy alone.

Example 3. A patient treated with radiotherapy 15 years ago develops a sarcoma within the treated field. What kind of effect is this, and what does it imply about dose?

A stochastic effect, specifically radiation-induced carcinogenesis.

Stochastic effects differ fundamentally from deterministic ones. They have no threshold dose, so no exposure can be declared entirely safe, and dose influences the probability that the effect occurs rather than its severity. A sarcoma arising after radiotherapy is no worse for having followed a higher dose; the higher dose simply made it more likely.

Deterministic effects behave in the opposite way. Cataract, skin erythema, sterility and acute radiation syndrome all require a threshold dose to occur at all, and above that threshold severity rises with dose.

Two clinical implications follow. Latency for second malignancy is long, typically a decade or more, which is why the risk weighs most heavily in children and young adults with long expected survival and matters least in elderly patients with limited life expectancy. And it is one of the principal reasons every technical advance in the field has aimed at reducing the volume of normal tissue irradiated rather than at delivering more dose to the tumour.

Example 4. Why is brachytherapy so central to the treatment of cervical cancer?

Because the geometry of the disease matches the physics of the technique better than any other common cancer.

Dose from a brachytherapy source falls off according to the inverse square of distance, so intensity drops extremely steeply over a few centimetres. Placing the source within the uterine cavity and the vaginal fornices therefore delivers a very high dose to the central tumour while the rectum and bladder, lying a short distance away, receive substantially less.

External beam radiotherapy alone cannot achieve this. To deliver an equivalent central dose it would have to pass that dose through the bladder and rectum, whose tolerance would be exceeded well before the tumour dose was reached.

The clinical evidence is unambiguous. Omitting brachytherapy in locally advanced cervical cancer is associated with markedly worse local control and survival, and it cannot be adequately substituted by additional external beam dose.

This has particular significance in India, where cervical cancer is a leading cause of cancer death in women and where access to brachytherapy facilities is a genuine determinant of outcome.

Example 5. A patient with widespread metastatic disease has a painful bone metastasis. The team proposes a six-week course of daily radiotherapy. Comment.

The fractionation is wrong for the intention.

The purpose here is palliative rather than curative, and the reasoning changes accordingly. Late effects, which are driven by dose per fraction and appear over years, matter far less in a patient whose survival is measured in months, so the constraint that forces small daily fractions in radical treatment does not apply.

The burden of treatment matters a great deal. Six weeks of daily attendance for a patient with limited mobility, pain and a short life expectancy consumes a substantial proportion of the time they have.

The evidence supports short courses. A single fraction is as effective as multi-fraction regimens for pain relief from uncomplicated bone metastases, with a somewhat higher retreatment rate that is easily managed by retreating.

Longer courses remain appropriate in specific situations, such as impending pathological fracture or spinal cord compression where a different technique and dose apply, but for uncomplicated painful bone metastasis a single fraction is the appropriate offer.

Summary

Radiation is not selective; differential repair between fractions creates the therapeutic ratio.

About two-thirds of X-ray damage is indirect, through free radicals from water.

The double-strand break is the lethal lesion.

Oxygen fixes damage, making well-oxygenated cells two to three times more sensitive.

Hypoxic tumour cores are radioresistant, and anaemia worsens outcomes.

The five Rs are repair, reassortment, reoxygenation, repopulation and radiosensitivity.

Repair of sublethal damage is the principal reason fractionation works.

Accelerated repopulation means gaps and prolonged treatment reduce cure rates.

Acute effects occur in rapidly dividing tissue and usually recover.

Late effects occur in slowly dividing tissue and are usually permanent.

Late effects depend mainly on dose per fraction rather than total dose.

Second malignancy has the longest latency and matters most in the young.

Megavoltage beams are skin-sparing because maximum dose is deposited below the surface.

Brachytherapy exploits inverse square fall-off to spare adjacent tissue.

Brachytherapy is essential in cervical cancer and cannot be substituted.

Conformal, intensity-modulated and image-guided techniques all spare normal tissue.

Concurrent chemoradiation is synergistic in both efficacy and toxicity.

Palliative treatment uses fewer, larger fractions, and one fraction often suffices for bone pain.

Radiotherapy is emergency treatment for cord compression, superior vena cava obstruction and tumour bleeding.

Deterministic effects have a threshold and increase in severity; stochastic effects have none and increase in probability.