Regression Analysis of Rectal Cancer and Possible Application of Artificial Intelligence (AI) Utilization in Radiotherapy

Artificial Intelligence (AI) has been widely employed in the medical field in recent years in such areas as image segmentation, medical image registration, and computer-aided detection. This study explores one application of using AI in adaptive radiation therapy treatment planning by predicting the tumor volume reduction rate (TVRR). Cone beam computed tomography (CBCT) scans of twenty rectal cancer patients were collected to observe the change in tumor volume over the course of a standard five-week radiotherapy treatment. In addition to treatment volume, patient data including patient age, gender, weight, number of treatment fractions, and dose per fraction were also collected. Application of a stepwise regression model showed that age, dose per fraction and weight were the best predictors for tumor volume reduction rate.

Variability of α/β Ratios for Prostate Cancer With the Fractionation Schedule: Caution Against Using the Linear-Quadratic Model for Hypofractionated Radiotherapy

Background: Prostate cancer (PCa) is known to be suitable for hypofractionated radiotherapy due to the very low α/β ratio (about 1.5-3 Gy). However, several randomized controlled trials have not shown the superiority of hypofractionated radiotherapy over conventionally fractionated radiotherapy. Besides, in vivo and in vitro experimental results show that the linear-quadratic (LQ) model may not be appropriate for hypofractionated radiotherapy, and we guess it may be due to the influence of fractionation schedules on the α/β ratio. Therefore, this study attempted to estimate the α/β ratio in different fractionation schedules and evaluate the applicability of the LQ model in hypofractionated radiotherapy. Methods: The maximum likelihood principle in mathematical statistics was used to fit the parameters: k, α and β values in the tumor control probability (TCP) formula derived from the LQ model. In addition, the fitting results were substituted into the original TCP formula to calculate 5-year biochemical relapse-free survival for further verification.Results: Information necessary for fitting could be extracted from a total of 23,281 PCa patients. A total of 16,442 PCa patients were grouped according to fractionation schedules. We found that, for patients who received conventionally fractionated radiotherapy, moderately hypofractionated radiotherapy, and stereotactic body radiotherapy, the average α/β ratios were 1.78 Gy (95% CI: 1.59-1.98, P < 0.001), 3.46 Gy (95% CI: 3.08-3.83, P < 0.001), and 4.24 Gy (95% CI: 4.10-4.39, P < 0.001), respectively. Hence, the calculated α/β ratios for PCa tended to become higher when the dose per fraction increased. Among all PCa patients, 14,641 could be grouped according to the risks of PCa in patients receiving radiotherapy with different fractionation schedules. The results showed that as the risk increased, the k and α values decreased, indicating that the number of effective target cells decreased and the radioresistance increased.Conclusions: The LQ model appeared to be inappropriate for high doses per fraction owing to α/β ratios tending to become higher when the dose per fraction increased. Therefore, to convert the conventionally fractionated radiation doses to equivalent high doses per fraction using the standard LQ model, a higher α/β ratio should be used for calculation.

Tumor Control Probability Modeling for Radiation Therapy of Keratinocyte Carcinoma

SummarySkin cancer patients may be treated definitively using radiation therapy (RT) with electrons, kilovoltage, or megavoltage photons depending on tumor stage and invasiveness. This study modeled tumor control probability (TCP) based on the pooled clinical outcome data of RT for primary basal and cutaneous squamous cell carcinomas (BCC and cSCC, respectively). Four TCP models were developed and found to be potentially useful in developing optimal treatment schemes based on recommended ASTRO 2020 Skin Consensus Guidelines for primary, keratinocyte carcinomas (i.e. BCC and cSCC).BackgroundRadiotherapy (RT) with electrons or photon beams is an excellent primary treatment option for keratinocyte carcinoma (KC), particularly for non-surgical candidates. Our objective is to model tumor control probability (TCP) based on the pooled clinical data of primary basal and cutaneous squamous cell carcinomas (BCC and cSCC, respectively) in order to optimize treatment schemes.MethodsPublished reports citing crude estimates of tumor control for primary KCs of the head by tumor size (diameter: ≤2 cm and &gt;2 cm) were considered in our study. A TCP model based on a sigmoidal function of biological effective dose (BED) was proposed. Three-parameter TCP models were generated for BCCs ≤2 cm, BCCs &gt;2cm, cSCCs ≤2 cm, and cSCCs &gt;2 cm. Equivalent fractionation schemes were estimated based on the TCP model and appropriate parameters.ResultsTCP model parameters for both BCC and cSCC for tumor sizes ≤2 cm and &gt;2cm were obtained. For BCC, the model parameters were found to be TD50 = 56.62 ± 6.18 × 10-3 Gy, k = 0.14 ± 2.31 × 10−2 Gy−1 and L = 0.97 ± 4.99 × 10−3 and TD50 = 55.78 ± 0.19 Gy, k = 1.53 ± 0.20 Gy−1 and L = 0.94 ± 3.72 × 10−3 for tumor sizes of ≤2 cm and &gt;2 cm, respectively. For SCC the model parameters were found to be TD50 = 56.81 ± 19.40 × 104 Gy, k = 0.13 ± 7.92 × 104 Gy−1 and L = 0.96 ± 1.31 × 10-2 and TD50 = 58.44 ± 0.30 Gy, k = 2.30 ± 0.43 Gy−1 and L = 0.91± 1.22 × 10−2 for tumors ≤2cm and &gt;2 cm, respectively. The TCP model with the derived parameters predicts that radiation regimens with higher doses, such as increasing the number of fractions and/or dose per fraction, lead to higher TCP, especially for KCs &gt;2 cm in size.ConclusionFour TCP models for primary KCs were developed based on pooled clinical data that may be used to further test the recommended kV and MV x-ray and electron RT regimens from the 2020 ASTRO guidelines. Increasing both number of fractions and dose per fraction may have clinically significant effects on tumor control for tumors &gt;2 cm in size for both BCC and cSCC.

Avasopasem manganese synergizes with hypofractionated radiation to ablate tumors through the generation of hydrogen peroxide

Avasopasem manganese (AVA or GC4419), a selective superoxide dismutase mimetic, is in a phase 3 clinical trial (NCT03689712) as a mitigator of radiation-induced mucositis in head and neck cancer based on its superoxide scavenging activity. We tested whether AVA synergized with radiation via the generation of hydrogen peroxide, the product of superoxide dismutation, to target tumor cells in preclinical xenograft models of non–small cell lung cancer (NSCLC), head and neck squamous cell carcinoma, and pancreatic ductal adenocarcinoma. Treatment synergy with AVA and high dose per fraction radiation occurred when mice were given AVA once before tumor irradiation and further increased when AVA was given before and for 4 days after radiation, supporting a role for oxidative metabolism. This synergy was abrogated by conditional overexpression of catalase in the tumors. In addition, in vitro NSCLC and mammary adenocarcinoma models showed that AVA increased intracellular hydrogen peroxide concentrations and buthionine sulfoximine– and auranofin-induced inhibition of glutathione- and thioredoxin-dependent hydrogen peroxide metabolism selectively enhanced AVA-induced killing of cancer cells compared to normal cells. Gene expression in irradiated tumors treated with AVA suggested that increased inflammatory, TNFα, and apoptosis signaling also contributed to treatment synergy. These results support the hypothesis that AVA, although reducing radiotherapy damage to normal tissues, acts synergistically only with high dose per fraction radiation regimens analogous to stereotactic ablative body radiotherapy against tumors by a hydrogen peroxide–dependent mechanism. This tumoricidal synergy is now being tested in a phase I-II clinical trial in humans (NCT03340974).

Reinforcement Learning for Radiotherapy Dose Fractioning Automation

External beam radiotherapy cancer treatment aims to deliver dose fractions to slowly destroy a tumor while avoiding severe side effects in surrounding healthy tissues. To automate the dose fraction schedules, this paper investigates how deep reinforcement learning approaches (based on deep Q network and deep deterministic policy gradient) can learn from a model of a mixture of tumor and healthy cells. A 2D tumor growth simulation is used to simulate radiation effects on tissues and thus training an agent to automatically optimize dose fractionation. Results show that initiating treatment with large dose per fraction, and then gradually reducing it, is preferred to the standard approach of using a constant dose per fraction.

Further development of spinal cord retreatment dose estimation: including radiotherapy with protons and light ions

A new graphical user interface (GUI) was developed to aid in the assessment of changes in the radiation tolerance of spinal cord/similar central nervous system tissues with time between two treatment courses. The GUI allows any combination of photons, protons (or ions) to be used as the initial, or retreatment, course. Allowances for clinical circumstances, of reduced tolerance, can also be made. The radiobiological model was published previously and has been incorporated with additional checks and safety features, to be as conservative as possible. The proton option includes use of a fixed RBE of 1.1 (set as the default), or a variable RBE, the latter depending on the proton linear energy transfer (LET) for organs at risk. This second LET-based approach can also be used for ions, by changing the LET parameters. GUI screenshots are used to show the input and output parameters for clinical situations used in worked examples from previous publications, where the proton and ion treatments required additional ‘longhand’ calculations. The results from the GUI are in agreement with previously published calculations, but the results are now rapidly available without tedious and error-prone manual computations. The software outputs provide a maximum dose limit boundary, which should not be exceeded. Clinicians may also choose a lower number of treatment fractions, whilst using the same dose per fraction (or conversely a lower dose per fraction but with the same number of fractions) in order to achieve the intended clinical benefit. The new GUI will allow rational estimations of time related radiation tolerance changes in the spinal cord and similar central nervous tissues (optic chiasm, brainstem), which can be used to guide the choice of retreatment dose fractionation schedules.

Dose compensation based on biological effectiveness due to interruption time for photon radiation therapy

Objective: To evaluate the biological effectiveness of dose associated with interruption time; and propose the dose compensation method based on biological effectiveness when an interruption occurs during photon radiation therapy. Methods: The lineal energy distribution for human salivary gland tumor was calculated by Monte Carlo simulation using a photon beam. The biological dose (Dbio) was estimated using the microdosimetric kinetic model. The dose compensating factor with the physical dose for the difference of the Dbio with and without interruption (Δ) was derived. The interruption time (τ) was varied to 0.1, 0.2, 0.3, 0.4, 0.5, 1, 2, 3, 4, 5, 10, 20, 30, 40, 50, 75, and 120 min. The dose per fraction and dose rate varied from 2 to 8 Gy and 0.1 to 24 Gy/min, respectively. Results: The maximum Δ with 1 Gy/min occurred when the interruption occurred at half the dose. The Δ with 1 Gy/min at half of the dose was over 3% for τ >= 20 min for 2 Gy, τ = 10 min for 5 Gy, and τ = 10 min for 8 Gy. The maximum difference of the Δ due to the dose rate was within 3% for 2 and 5 Gy, and achieving values of 4.0% for 8 Gy. The dose compensating factor was larger with a high dose per fraction and high-dose rate beams. Conclusion: A loss of biological effectiveness occurs due to interruption. Our proposal method could correct for the unexpected decrease of the biological effectiveness caused by interruption time. Advances in knowledge: For photon radiotherapy, the interruption causes the sublethal damage repair. The current study proposed the dose compensation method for the decrease of the biological effect by the interruption.