Skip to main content
Radiotherapy

Radiotherapy

Tumour growth models shed light on radionuclide therapy

06 Mar 2019 Tami Freeman
The researchers
The researchers (left to right): Francisco Liberal, Joe O'Sullivan, Stephen McMahon and Kevin Prise. (Courtesy: Stephen McMahon)

Prostate cancer is the most common cancer in men and at later stages of the disease many patients develop painful bone metastases. One promising modality for management of skeletal metastases is targeted radionuclide therapy, in which a radioactive drug travels through the patient’s bloodstream to the tumour where it delivers radiation to cancer cells.

Radium-223, which targets areas of increased bone turnover, has emerged as a key radionuclide for such treatments. The atoms decay via emission of alpha particles that deposit a high amount of energy over a short distance (70–100 μm), damaging the DNA of targeted cells while limiting exposure to healthy tissue.

A large trial (ALSYMPCA) of radium-223 dichloride (223RaCl2) in patients with metastatic prostate cancer bone metastases provided proof that the treatment prolonged the time to a patient’s first symptomatic skeletal event (SSE) and increased overall survival. But questions remain regarding the dosimetry and pharmacodynamics of 223RaCl2.

To investigate the mechanisms of action of 223RaCl2, a team at Queen’s University Belfast has performed mathematical modelling of tumour growth using three different uptake models. To determine the most realistic scenario, they compared the models’ predictions of time to first SSE with published clinical data (J. Radiat. Oncol. Biol. Phys. 10.1016/j.ijrobp.2018.12.015).

Three scenarios

The researchers used the established Gompertz model to simulate tumour growth, and incorporated the effects of 223Ra treatment into the model, based on three biophysical scenarios. First, they considered uniform exposure, which assumes that all tumour cells are equally affected by radiation dose.

“We were surprised how much the uniform model over-estimated the effects of 223Ra treatment, so we knew that less of the tumour must be exposed,” explains co-author Stephen McMahon.

As such, McMahon and colleagues tested two other scenarios: an outer layer effect, where only the surface of the metastatic volume is exposed to radiation; and constant volume exposure, where the number of affected cells remains constant throughout tumour growth.

Three exposure scenarios

“With suitable dose-rate tuning, we were able to get good agreement for patients who failed at late time-points. This suggested that the uniform model was good for small tumours, but rapidly saturated — and the constant volume model is the simplest approximation of that behaviour,” says McMahon. “The true biology is likely more complex, but this worked well for this initial analysis.”

To test their models, the researchers employed clinical data from the aforementioned  223RaCl2 trial, which included both treated and control groups. They used these data to generate a “virtual patient population” with a range of initial tumour volumes that reproduced the observed time to SSE in the control group.

The researchers then simulated the effects of 223Ra treatment on the virtual population, using each of the three radiation exposure models. To relate tumour growth to SSE, they assumed that skeletal events occurred when the number of tumour cells reached 80% of the total tumour burden that can be supported by the patient.

 Clinical comparisons

Using an initial dose rate based on published rates for an average-sized patient, the uniform effect scenario produced over-optimistic results. It gave 12- and 6-month delays in reaching the number of cells corresponding to an SSE, for late and early tumour stages, respectively. Even with a lower initial dose rate, this model overestimated the effects on patients with high disease burden and under-estimated the effects on patients with lower disease burden.

Data comparison

The outer layer effect scenario predicted cell killing rates that were too high for early tumour growth stages and too low for later tumour growth stages, with poor agreement with clinical data.

The constant volume exposure scenario provided the closest match to the clinical data for patients treated with 223RaCl2. These results suggest that the effects of 223Ra saturate rapidly with tumour volume, only affecting a constant number of cells regardless of tumour growth (once the tumour volume exceeds a certain limit).

The authors conclude that metastatic tumour cells do not experience a uniform dose exposure, with only a sub-population of the tumour affected by 223Ra. This finding is particularly significant since uniform distribution of radionuclide activity in bone metastatic volumes is frequently assumed in conventional dosimetric calculations.

“We were able to show that some common models of 223Ra uptake did not work well to describe the clinical data, and that there was strong evidence of saturation of uptake in even relatively small metastases,” says McMahon. “This represents some of the first biophysical analyses of these data, and the saturating uptake, if verified, may have significant implications for future attempts to optimize drug design and scheduling in radionuclide therapies.”

McMahon notes that the team’s clinical partners are currently completing a trial combining 223Ra treatment with external-beam radiotherapy. “This is an exciting opportunity, as this trial includes detailed molecular and MR imaging, which will enable us to quantify the disease burden and delivered dose to these patients in detail,” he tells Physics World. “This will give us a valuable testing data set to validate the model’s assumptions.”

Copyright © 2024 by IOP Publishing Ltd and individual contributors