Radiation damage, repopulation and cell recovery analysis ofin vitrotumour cell megacolony culture data using a non-Poissonian cell repopulation TCP model

Analysis of a cohort of prostate patients treated with HDR mono-brachytherapy

The aim of this study is to perform volumetric and basic radiobiological analyses using the database on prostate patients treated by HDR brachytherapy in our institution during the period 2011-2016. Real-time ultrasound based technique was used, with Oncentra Prostate planning software. The whole period was divided into two sub-periods, according to the 100% dose per fraction, which was 10.5 Gy during the first period (2011-2012), and 11 Gy during the second period (2013-2016), for each of the three fractions. The follow up time varied from 19 to 81 months, with a median of 45 months and a mean of 47 months. The uniformity of the treatment technique for both periods is investigated. Tumour Control Probability (TCP) values for the expected local control are calculated according to a population phenomenological TCP model for different values of the α/β ratio. The calculations are based on the obtained averaged Dose Volume Histograms for the two investigated sub-periods. 74 patients were treated in total. Local control failure is observed in 5 cases, which corresponds to an observed TCP = 93.2%. The comparison of the calculated population average DVH with the DVHs of the cases with local control failure shows that in 4 of them, doses higher than average were delivered to the prostate. It is shown that the uniformity of the treatment was improved during the second sub-period. A possible explanation of the observed failures may be that these cases exhibit inherent tumour cell radio-resistance higher than average. Our radiobiological analysis indicates a α/β ratio value somewhat higher than the one currently accepted. The value of the prostate α/β ratio is estimated to be in the range of [3.5-6] Gy.

From cell population models to tumor control probability: including cell cycle effects

BACKGROUND

Classical expressions for the tumor control probability (TCP) are based on models for the survival fraction of cancer cells after radiation treatment. We focus on the derivation of expressions for TCP from dynamic cell population models. In particular, we derive a TCP formula for a generalized cell population model that includes the cell cycle by considering a compartment of actively proliferating cells and a compartment of quiescent cells, with the quiescent cells being less sensitive to radiation than the actively proliferating cells.

METHODS

We generalize previously derived TCP formulas of Zaider and Minerbo and of Dawson and Hillen to derive a TCP formula from our cell population model. We then use six prostate cancer treatment protocols as a case study to show how our TCP formula works and how the cell cycle affects the tumor treatment.

RESULTS

The TCP formulas of Zaider-Minerbo and of Dawson-Hillen are special cases of the TCP formula presented here. The former one represents the case with no quiescent cells while the latter one assumes that all newly born cells enter a quiescent cell phase before becoming active. From our case study, we observe that inclusion of the cell cycle lowers the TCP.

CONCLUSION

The cell cycle can be understood as the sequestration of cells in the quiescent compartment, where they are less sensitive to radiation. We suggest that our model can be used in combination with synchronization methods to optimize treatment timing.

Examining the validity of Poissonian models against the birth and death TCP model for various radiotherapy fractionation schemes

PURPOSE

To compare the results obtained for tumour control probability (TCP) in protracted treatments, we used two models which apply Poisson statistics for clonogenic cell distribution and a non-Poissonian model, emphasising the conditions for the validity of Poissonian models. Previously published results on two cell lines growing as megacolonies in vitro irradiated with conventional and accelerated dose fractionation schemes were used.

MATERIALS AND METHODS

The expressions of TCP for three models are described and conclusions are drawn on the applicability of each model, and their usefulness for different fractionations.

RESULTS

The fits to experimental data are shown and the parameter values for both Poissonian and non-Poissonian models are given. We also determined if differences exist in repopulation rate and other related parameters, for different protocols of treatment.

CONCLUSIONS

Both the Poissonian models, when they satisfied the required conditions, and the non-Poissonian model, gave acceptable fits. We observed no significant differences in repopulation for different irradiation protocols.

Optimization of the temporal pattern of radiation: An IMRT based study

PURPOSE

To investigate how the temporal pattern of dose applied during a single-intensity modulated radiation therapy (IMRT) fraction can be arranged to maximize or minimize cell kill.

METHODS AND MATERIALS

Using the linear-quadratic repair-time model and a simplified IMRT delivery pattern model, the surviving fraction of cells for a single fraction was calculated for all permutations of the dose delivery pattern for an array of clinically based IMRT cases. Maximization of cell kill was achieved by concentrating the highest doses in the middle of a fraction, while minimization was achieved by spreading the highest doses between the beginning and end. The percent difference between maximum and minimum cell kill (%Diff(min/max)) and the difference between maximum and minimum total doses normalized to 2 Gy/fx (deltaNTD(2 Gy)) was calculated for varying fraction durations (T), alpha/beta ratios, and doses/fx.

RESULTS

%Diff(min/max) and deltaNTD(2 Gy) both increased with increasing T and with decreasing alpha/beta. The largest increases occurred with dose/fx. With alpha/beta = 3 Gy and 30 min/fx, %Diff(min/max) ranged from 2.7-5.3% for 2 Gy/fx to 48.6-74.1% for 10 Gy/fx, whereas deltaNTD(2 Gy) ranged from 1.2 Gy-2.4 Gy for 30 fractions of 2 Gy/fx to 2.3-4.8 Gy for 2 fractions of 10.84 Gy/fx. Using alpha/beta = 1.5 Gy, an analysis of prostate hypofractionation schemes yielded differences in clinical outcome based on the pattern of applied dose ranging from 3.2%-6.1% of the treated population.

CONCLUSIONS

Rearrangement of the temporal pattern of dose for a single IMRT fraction could be used to optimize cell kill and to directly, though modestly, affect treatment outcome.