A New Era in Biologically-Targeted Radiation Therapy
Accelerator-based neutron sources and novel boron-10 target drug development are propelling the resurgence in BNCT interest globally. BNCT is the binary therapy modality that combines neutron radiation and boron-10 targeted drugs, such as BPA (boronophenylalanine). Because of the unique nature of cell-selective radiation, BNCT is the therapy with the lowest impact on the patient’s quality of life compared to other radiation, chemotherapy, or biological therapy modalities, as it selectively destroys cancer cells and requires at most two fractions.
One historical challenge in elevating BNCT to a standard of care is the difficulty in accessing safe and effective neutrons for clinical use. Now, the availability of compact neutron technology allows for installations in existing or new hospital facilities utilizing a small footprint—offering new promise for the broad adoption and application of BNCT as a truly biologically targeted cancer treatment.
Click any topic below to see the Q&A
Q1: What cancers are currently being treated in the clinic using BNCT?
A: BNCT has mainly been performed in phase I/II, difficult-to-treat malignancies including glioblastoma multiforme, head and neck cancer, multifocal hepatocellular carcinoma, recurrent lung cancer, squamous cell carcinomas, salivary gland carcinomas, sarcomas, recurrent malignant meningioma, angiosarcoma and extramammary Paget's disease. These particularly aggressive and difficult-to-treat cancers, along with nearly all head and neck cancers, present unique challenges in treatment. Yet where conventional treatments hit hard limits in effectiveness, biologically targeted treatments like boron neutron capture therapy (BNCT) deliver exciting promise.
Global experience in patient treatments is detailed below:
|Cancer Type||# of Patients||Country||Outcomes|
|Current Clinical Treatments|
|Recurrent head and neck (accelerator-based)||100+||Japan||Hirose: 21 patients|
|Recurrent head and neck (reactor-based)||88||Japan||Kato et al, 26 patients Median survival: 13.6 months; 62 patients, Median survival: 10.1 months|
|Head and neck (definitive)||30||Finland||Response rate: 76%|
Median PSF: 7.5 months; 2-yrs OS=30%
|Recurrent head and neck||79||Finland||Complete response rate: 36%, 2 yr OS 21%|
|Recurrent head and neck||23||Taiwan||2-yr local regional control: 28%; 2 yr OS 47%|
|Glioblastoma (definitive)||96||Japan||Yamamoto et al: 15 patients median OS: 25.7 months|
Kageji et al: 23 patients median OS: 19.5 months
Miyatake et al 58 patients: Median OS: 23.5-27.1 months
|Glioblastoma (recurrent)||46||Japan||Kawabata et al: 24 patients median OS: 18.9 months|
Miyatake et al: 22 patients, Median OS: 10.8 months (recurrent)
|Recurrent Glioblastoma||30||Finland||Median OS: 14.2 months|
|Recurrent malignant meningioma||19||Japan||Median OS: 14.1 months|
|Recurrent malignant meningioma||30||Finland|
|Cutaneous Melanoma||7||Argentina||Response rate: 69% and OS 23 months post BNCT|
|Cutaneous Melanoma||8||Japan||5 patients alive with no evidence of disease for 5.6 to 8.2 years after BNCT (2020)|
Q2: How many clinical BNCT treatments have been delivered worldwide to date? And are there systems currently installed? Where?
A: There is a substantial global market opportunity for new cancer treatment modalities. The statistics of half a million new patients annually in Europe, China, Japan, and the US show a market potential for BNCT centers to be clinically operating globally.
Currently, approximately 2000 patients have been treated worldwide with BNCT, and there are about sixteen BNCT centers with compact accelerator-based BNCT systems in various stages of development around the world.
Today, Japan has been the leader in clinical BNCT, with years of accomplished clinical trials and successful operation of several fully-functioning clinics. In 2020, the Japanese national health insurance system approved coverage of BNCT treatments for recurrent head and neck cancer treatment and currently efforts are being made to include coverage for brain cancer.
China and South Korea have also built BNCT centers and have started treating patients under clinical trials for recurrent head and neck and glioblastomas.
Europe has a new accelerator-based BNCT center at Helsinki University, who aim to start clinical trials later in 2023. Other BNCT facilities are being built including Italy (CNAO) or in planning phase including the United Kingdom, Spain, and Poland.
Q1: How many patient treatments can be done in a day with BNCT?
Q2: What does the BNCT workflow look like?
A: The current clinical workflow for accelerator based BNCT with the available BPA drugs starts with pre-planning based on the available CT or MRI. The latter is used to perform pre-calculation to verify whether BNCT is indicated for the patient. If it is, the next step is preparation stage, simulating patient positioning and imaging in several steps. One to two weeks before treatment, the team performs a PET CT scan of the patient—called an F-18 labelled boronophenylalanine PET, or “F-BPA PET—so they make sure the patient can be treated with BNCT. F-BPA PET shows the boron uptake in the patient’s tumor before they can be treated. If the uptake is sufficient, the team moves to the next step, patient-positioning. This is usually performed one week before the actual treatment session.
The patient positioning step is similar to the X-ray verification for conventional radiation therapy, for example for head and neck treatments, technicians use a thermoplastic mask to immobilize the patient’s positioning. The team also performs a CT scan which is to be used for the final treatment planning process. In BNCT, particularly for head and neck treatment, it is often challenging to acquire the correct angles for specific patients. For example, the Osaka BNCT facility in Japan uses two couch orientations, depending on the location of the patient’s tumor – either the supine on the couch for brain cancer or the couch chair for head and neck cancer.
On the day of the treatment, a technician initiates the boron-10 infusion for the patient 90 minutes to two hours before irradiation. After up to an hour has passed, the team positions the patient, and a radiology technician performs X-ray imaging to verify the setup. Immediately before targeting with neutrons, the team takes a final blood sample to confirm the blood-boron concentration that determines the radiation time. After the BNCT treatment, the patient returns to the BNCT center generally after 24 to 48 hours for a check-up.
Q1: What is the evidence that, in humans, Boron accumulates selectively in the cancer?
A: There are several levels of evidence for tumor specific localization of boronated drugs used in BNCT:
- First, no tumor response would be observed with the neutrons used at this energy unless there was a stopping reaction, which requires Boron. Therefore, all observed responses de facto suggest tumor localization of Boron.
- As to specificity of localization, the most recent data are the F-BPA PET images which visually demonstrate tumor-specific Boron localization (see example image below). In China, patients are selected for treatment based on PET scan positivity using F-BPA PET.
- Mechanistically, the surface transporter, LAT1, is associated with transport of BPA into tumor cells, with the observation that there is significant upregulation of LAT in several cancers, especially recurrent head and neck cancer (as well as other malignancies). The following image is an example of LAT-1 expression in head and neck and triple-negative breast cancer tumor samples that were analyzed by TAE Life Sciences
Q2: Is there any scatter radiation or systemic adverse effects to OARs? If so, what is the effects profile?
A: Side-effects or systemic adverse effects on OARs from BNCT are a function of at least 4 phenomena:
- A small amount of gamma irradiation that results from the primary reaction. The vast majority of resultant radiation is alpha, hence high LET but with very shallow penetrance. A small amount of incidental gamma could in theory contribute to some toxicity.
- Beam contamination by fast neutrons is a function of the beam production system and these neutrons have the capability to interact with tissue without requiring Boron and account for some of the side effects. In some of the Japanese clinic’s experiences, the prescribed tumor dose is arrived at based on identifying tolerable skin toxicity as the dose-selection point.
- Based on the drug used, there is some degree of non-specific Boron uptake in normal tissues such as skin, mucosa, etc. This is possibly the largest contributor to OAR toxicities, and varies by tissue and drug.
- Yet another mechanism of toxicity is the rapidity with which tumor responses occur. In the intracranial context, this can lead to pseudo progression which needs to be managed. In the recurrent GBM setting, Bevacizumab is frequently used in combination to avoid this.
Q1: How have patients been selected for BNCT—especially the non-recurrent cases? Is there any potential bias in this selection criteria?
A: Since most studies are conducted at the institutional level, one would expect certain biases. In China, the selection is based on positivity of F-BPA PET. In Japan, the experience has largely been in recurrent refractory cancers, and therefore patient selection reflects very advanced patients. Both in Japan and in Finland, small clinical trials with specific eligibility factors have been designed and conducted and, in these trials, the eligibility criteria drive patient selection.
Therefore, there is surely bias in patient selection. Usually, the cases have involved treating large, unresectable, often inoperable, recurrent tumors.
Q2: Is BNCT suitable for all cancer patients?
A: BNCT may be utilized for several cancers that are resistant to many, if not all,
currently available cancer therapies. BNCT has mainly been performed in phase I/II, difficult-to-treat malignancies including recurrent gliomas, recurrent head and neck cancer and metastatic melanoma.
Many years of global research has demonstrated BNCT to be safe and effective in treating very advanced and untreatable cancers, as well as increasing survival rates in patients. With advances in the development of new boron compounds, there is interest to begin treating patients with other, traditionally treatable cancers such as breast, lung, colorectal and other forms of cancer in clinical trials.
Q3: How do you verify the concentration of boron drugs in-vivo (during treatment)?
A: Blood is drawn at one or more points during the infusion of the boron drug on the day of treatment, and the boron concentration is measured at these time points (in units of parts-per-million or micrograms boron-10 per gram of tissue). A standardized tumor-to-blood uptake ratio is then used to predict the tumor concentration of boron-10, and the required beam-on time is calculated.
More precisely, the existing treatment plan assumes a concentration of boron-10 in the tumor, with a calculated beam-on time to deliver the desired tumor dose; this beam-on duration is adjusted before treatment to reflect the measured blood boron-10 concentration.
Research is taking place to find alternative ways to measure the actual boron-10 concentration in the tumor at time of treatment. Prompt gamma imaging of the characteristic 480-keV gamma ray that is produced in most boron-10 neutron capture reactions is the most mature approach, although it is still not yet able to quantitatively predict tumor boron concentrations.
Q1: What is the necessary proton current to achieve the required neutron beam intensity for clinical use?
A: To deliver tumoricidal dose with boronophenylalanine (BPA) in one hour or less, an epithermal neutron flux exiting the beam shaping assembly (BSA) of at least 5 x 108 n/cm2-sec may be necessary. In general, increasing the epithermal flux will reduce the treatment time. There is not a one-to-one correlation between the intensity of the proton beam striking the target and the epithermal neutron flux that exits the BSA – the choice of moderator and reflector materials in the BSA, as well as the geometry of the components, strongly impact the conversion efficiency of incident protons to exiting epithermal neutrons.
Citation: INTERNATIONAL ATOMIC ENERGY AGENCY, Advances in Boron Neutron Capture Therapy, Non-serial Publications, IAEA, Vienna (2023)
Q2: Based on cell damage studies, many authors have suggested that the accelerator neutron beams should be much less damaging than the old reactor beams. Do you believe you are seeing evidence of that in your patients?
A: Skin damage from neutrons is primarily due to two reactions: elastic scattering with hydrogen, which is dominated by higher energy neutrons, and the (n,p) reaction with nitrogen-14, which is dominated by low energy (also called thermal) neutrons. While the dose imparted to tissue from the Nitrogen reaction depends on the flux of neutrons and concentration of nitrogen, the Hydrogen reaction also depends strongly on the energy of the reacting neutrons: with hydrogen, up to 100% of the neutron’s kinetic energy can be imparted to the hydrogen nucleus in a single collision. This means that on average, a 3-MeV neutron will deposit 300 times more dose than a 10-keV neutron (an ideal epithermal neutron energy for BNCT), and a 30-MeV neutron will deposit 3000 times more dose. Very small contributions of neutrons above 1 MeV can have a significant impact on the skin dose for this reason, and the energies and ranges of the recoil protons are also predicted to have higher relative biological effectiveness from hydrogen scattering (RBE-H). With lower contamination percentages of high energy neutrons in accelerator beams compared to reactor beams, the overall RBE-H is expected to be lower.
There is not yet sufficient clinical data to confirm the lower RBE-H for accelerator beams, but cell survival curves have estimated the following values are expected (compared to a historic RBE-H of 3.0-3.2 for reactor neutron beams):
- RBE-H for an accelerator-generated epithermal beam using a Beryllium target is 2.2-2.6
- RBE-H for an accelerator-generated epithermal beam using a Lithium target is 1.7-1.9
- RBE-H for a generic accelerator-generated epithermal beam is 2.0-2.5
- K. Ono, Prospects for the new era of boron neutron capture therapy and subjects for the future, Ther. Radiol. Oncol. 2 (2018).
- Y. Imamichi Sasaki, M. Ihara, T. Onodera, O. Chen, S. Nakamura, H. Okamoto, J. Itami, M. Masutani,Evaluation of the BNCT System in National Cancer Center Hospital Using Cells and Mice, The International Congress of Radiation Research, (2019).
- Anna J. Mason, Valerio Giusti, Stuart Green, Per Munck af Rosenschöld, T. Derek Beynon & John W. Hopewell (2011) Interaction between the biological effects of high- and low-LET radiation dose components in a mixed field exposure, International Journal of Radiation Biology, 87:12, 1162-1172, DOI: 10.3109/09553002.2011.624154
Q3: What's the diameter and thickness of Li layer on your target?
A: Neutrons are produced from proton bombardment of lithium targets via the 7Li(p,n)7Be reaction, which has a threshold energy of 1.89 MeV. As the incident proton beam enters the target, protons will begin to slow down, continuously changing the cross section (probability) of undergoing the (p,n) reaction. After the protons slow down below the 1.89 MeV threshold, no neutrons will be produced, although contamination photons are possible via other reactions with lower threshold energies. The resulting epithermal neutron beam exiting the BSA is typically collimated to a diameter greater than 10 cm, because sharp dose gradients in BNCT occur from differential boron uptake between tissue types, not because of radiation attenuation. For this reason, the proton beam striking the lithium target covers an area greater than 40 cm2, and the lithium thickness is greater than the range of protons slowing down from the incident proton energy to the threshold energy.
Q4: BNCT vs Proton and carbon ion therapy, what is the advantage?
A: BNCT is a high-LET, binary therapy modality that combines neutron radiation and boron-10 (10B) targeted drugs, such as BPA (boronophenylalanine) to biologically target cancer cells. Either component on its own does not produce a therapeutic radiation dose, but the combination releases an intense, highly localized dose with high relative biological effectiveness (RBE) around 3.0 that is deposited only in the cell that contains 10B. With selective 10B uptake in tumor cells, this permits cancer destruction on a cell-by-cell basis while sparing adjacent normal tissue cells. Proton therapy uses an RBE slightly higher than 1 (1.1 is commonly used for treatment planning), and Carbon Ion therapy is also high-LET and high-RBE (around 3.0), but both therapies require geometric targeting of the tumor with Bragg peaks.
Given BNCT uses biological targeting of cancer cells, it is less sensitive to tumor motion or changes in the patient’s anatomy. Other radiation therapies are limited by the resolution of medical imaging modalities and tumor targeting is only as good as the ability to get an accurate visual of the cancer cells’ location.
BNCT offers other unique advantages including:
- Accelerators with lithium targets use 2.5 – 2.6 MeV protons, up to 100 times lower than proton/carbon; this significantly reduces the power operational costs per patient.
- BNCT treatments are planned as single-fraction courses, sometimes with a second treatment delivered weeks later. This extreme hypofractionation helps support high patient throughput.
- The shielding thickness requirements are higher with proton and carbons vs BNCT, due to higher neutrons energies.
Both neutrons from reactor and accelerator may cause AE for healthy tissue equally. Only the energy decides it.
Adverse effects from neutron-based interactions depend on energy of the neutron and the element with which the neutron interacts, e.g. an elastic neutron scatter with hydrogen can produce different dose than the products of an (n,p) reaction with nitrogen-14. The source of the neutron is immaterial to the reaction probabilities for a given energy. The difference between reactor and accelerator neutron sources for BNCT is the relative contribution of different energies to the neutron spectrum. For a lithium target with 2.5-MeV protons, for example, the highest neutron energy possible is less than 1 MeV, while all reactor sources will have high energy tails extending up to about 10 MeV. At these energies, elastic scattering by hydrogen is the dominant reaction in tissue, and the dose deposition per interaction is proportional to the neutron energy. Great care is required to choose the combination of proton energy, target type and BSA design that creates a neutron energy spectrum that reduces risks of adverse events at the skin surface.
Q5: What beam arrangement would be used (e.g. for the nasopharynx example)?
A: The treatment plan example shown in the ESTRO seminar used a pair of parallel-opposed beams with a 15-cm diameter collimator. The separation between the collimator and patient surface is only 2 cm, which is typical for BNCT treatments: the rapid divergence of the neutron beam, even when well-collimated, requires patient placement close to the beam port. For this reason, the patient shoulders may be a limitation on the ability to achieve a true parallel-opposed beam arrangement, and a “near” parallel arrangement may be necessary with little change expected in the final isodose distributions and dose volume histogram.
Q6: What will TAE equipment do with IGRT? Also, is the lithium rotary type? Finally, what is the maximum proton current?
A: The Alphabeam system will use a combination of multiple imaging modalities in initial clinical trials to determine the sensitivity of BNCT treatments to patient setup uncertainties. This combination includes a portable CT scanner with reproducible positioning within the treatment room, which is removed before the neutron beam is turned on, as well as surface image guidance and wall-mounted lasers. BNCT will initially employ forward treatment planning, and therefore will not strictly be an IGRT-type of modality. The non-rotating lithium target is designed to operate up to 10 mA of proton current, and the matching BSA is designed to produce a corresponding epithermal neutron flux greater than 109 neutrons/cm2-sec.
Example 1: Nasopharyngeal tumor plans
- Large, central tumor abutting oral mucosa (Dose-limiting OAR: D0.01cc < 12 Gy-Eq)
- Monte Carlo dose calculation
- Parallel-opposed beams
- Compare 2 plans:
- Published BPA uptake in patients
- Scaled uptake based on mouse PK results
Example 2: Nasopharyngeal tumor plans
- New drug #1 vs BPA (assumptions based on mouse PK data):
- tumor 2.5x higher 10B concentration
- OAR 10B unchanged
- RBEs unchanged
- Note step change in dose at tumor/healthy tissue interface
- Increased 10B in tumor allows Rx based on tumor coverage goal
- Reduces OAR doses and treatment times
Q1: We can see the improvement of boron uptake in you slides—is there any improvement of the tumor-to-normal tissue (T/N) ratio?
A: Pharmacokinetic testing of next-generation BNCT compounds includes time-dependent measurement of boron uptake in both tumor and organs at risk (OARs). Improved therapeutic ratio beyond BPA is expected to require both higher boron uptake in tumor and greater tumor-to-normal tissue (T/N) ratios, for greater OAR sparing. BNCT dosimetry also utilizes biological effectiveness-weighted doses, in units of gray-equivalents, and testing of new compounds will require assessment of the new compound biological effectiveness (CBE) in both tumors and OARs. Small molecule candidate drugs developed by TAE Life Sciences have shown both increased tumor uptake and higher T/N ratios, compared to BPA.
Q2: Wouldn't a theranostic approach with imaging of the boron drug uptake make most sense for treatment planning?
A: BNCT’s unique combination of external radiation source and infused pharmaceutical suggests a natural application of the theranostic approach. If the drug used for BNCT treatment is radiolabeled with 18F, for example, it forms a theranostic pair with the original. There are limitations to this approach, however. The longstanding process to add 18F to BPA requires some formulation processes that are not available at all institutions. In addition, the time-dependent pharmacokinetics of both BPA and 18F-BPA have been shown to vary according to the speed of the infusion, e.g. a bolus injection for a PET scan will have different uptake kinetics than a 2-hour infusion for treatment. More research is required to find predictable and quantitative processes to take advantage of the theranostic potential of BPA and new BNCT drugs.
Q3: Did you try nude mice xenograft models with human cancer lines?
A: We will be using the FaDu hypopharyngeal xenograft model in nude mice later in 2023.
Q4: How did you formulate the boron compounds for TC440 and TC442?
A: Unlike BPA, which is formulated with fructose, our new compounds were formulated in aqueous solutions at 100 mg/kg or greater, which is 3 times more than the upper limit of BPA (70mg/kg)
Q5: What are the biological targets of the tumor with your new boron drugs?
Q1: What is the estimated cost of building out and operating a BNCT center?
A: The cost of a BNCT consists of the initial setup cost and the annual operating expenses. The setup cost of a BNCT center in the US is estimated to be $40 – $50 million USD. The latter covers the facility construction, the procurement and commissioning of the accelerator-based BNCT neutron system, and associated hardware and software.
- Device costs: A representative BNCT device costs $25 million and includes the accelerator-neutron source, the treatment planning software, robotic treatment couch, beam shaping assembly, imaging equipment, and other associated hardware and software.
- Facility setup costs: The facility costs 15 – 25 million USD depending on the number of treatment rooms and whether the facility is a stand-alone BNCT center or a repurposed existing facility.
- Operational costs: The annual operational costs including running and maintaining the BNCT system, the staff expenses, and other operational costs.
- Service & maintenance costs: The BNCT system annual service and maintenance agreement very between BNCT system vendors and may run $1.5 to $2.5 million per year.
- Staffing costs: The staffing costs vary based on medical center and would include the healthcare, administration and management services for the BNCT center.
Additional costs: Other costs that should be includes are marketing and daily operations of the center.