Radiobiology of LDR, HDR, PDR and VLDR Brachytherapy - GEC-ESTRO Handbook of Brachytherapy

Radiobiology of LDR, HDR, PDR and VLDR Brachytherapy

15

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part I: The Basics of Brachytherapy Version 1 - 22/10/2015

ern techniques of dose distribution optimization. Best examples are the treatment of cervix and uterine cancers where the blad- der, rectum, sigmoid and small bowel can be relatively spared, while the target is treated with relatively large fraction sizes doses (6-7Gy).HDR BT is a good alternative to LDR BT in cervix can- cer (Newman 1996, Orton 1991) as long as fraction sizes to point A do not exceed 7 Gy (which will result in fraction sizes of < 5 Gy to the rectum ICRU point. The same holds for prostate cancer (Kovacs 1999,Galalae2004, Martinez 2011, ,Hoskin 2012 A, Hoskin 2012 B, ,Zambouglou 2013))and breast cancer (Hammer,Resch) where - when treated with appropriate techniques - rectum ,urethra and breast skin can be spared, while the target receives significantly higher frac- tion sizes (7-10 Gy). The fact that both prostate cancer and breast cancer seem to have considerably higher repair capacity than squamous cell ca (respectively an α/β ratio around 1.5 for pros- tate cancer (Brenner and Hall, Fowler 2013) and 4.1 for breast cancer (Yarnold) further improves the therapeutic ratio when hypofractionation schedules are applied. Such a strategy is not possible for head and neck, anal canal or skin cancer when the target and the organ at risk are in the same volume. In principle, further fractionation would be needed to keep the therapeutic ratio of tumour cell kill to normal tissue damage acceptable. There are only a few series with small numbers that report on the use of HDR BT as treatment modality for head and neck, anal canal or skin cancer. This experience indicates that good local control (87-100 %) and acceptable complication rates (0- 15 % G2-G3) can be obtained in small series of patients with lip (Guinot 2003, Finestres 2005) or oral cavity cancer (Donath 1995, Leung 2002,Inoue 1996,In- oue 1998) , treated with HDR BT alone with doses of 45-60 Gy in fraction sizes of 4.5 to 6 Gy. In one series of oral cavity cancer (45) however treated to a lower dose (45.5) but higher fraction size (6.5 Gy), local control was only 53 % but the complication rate went up to 35%. So for HDR BT alone in head and neck cancer a total dose of 50-55 Gy in fraction sizes of 4.5 to 5.5 Gy seems to offer the best tradeoff. (Nag 2001, Mazeron 2009). The experience of HDR boost (10- 36 Gy in 3-6 Gy fractions) af- ter 45 Gy external beam RT in oropharyngeal cancer (Levendag 1997, Nose 2004) seems to lead to good local control rates (82-87 % but poor complication rates (29-30 %). So it seems advisable to limit total dose to 70- 75 Gy and fraction sizes of BT boost to 3-4.5Gy. 9.4 PDR clinical data Theoretically PDR could offer the best of both worlds, com- bining the advantages of mimicking the radiobiological effect of continuous LDR, with the technological advantages of HDR stepping source technology, full radioprotection and the possi- bility to optimize dose distribution, in 3D space as well as in time modifying pulse size and interval. In principle, to mimic the ef- fects of continuous LDR, the gaps between pulses should be not too long and hence pulse sizes not too large. This is because high-

er pulse sizes allow for less repair in between pulses and make a PDR schedule relatively more toxic to tissues with a large repair capacity and a fast half time of repair. The magnitude of this ef- fect was considered by Brenner and Hall (11) who concluded that for gaps between pulses up to 60 min, the radiobiological difference from continuous LDR could be an acceptable trade off to the gain in radioprotection and dose distribution control. The importance of small pulse sizes has been stressed, especially when irradiating cells with a lower α/β ratio and fast half-time of repair, when it was realised that PDR dose delivery by the step- ping source is like a golf ball, travelling through the treated area and delivering doses to individual cells from a few dwell posi- tions at a very high dose rate (Fowler 1997 2001). Since the first clinical experiments with technical feasibility (De Pree 1999, Peiffert) and early results (De Pree 1999, Levendag 1997, Swift1997) were published, there has been growing expe- rience in head and neck (,Strnad 2003,Strnad 2005, Johansson 2009, Johansson 2010) ,breast (Fritz 1997, Fritz 2000,Harms 2001,Harms 2002) , anal canal Gerard1999) and cervix cancer (Bachtiary 2005, Swift 1997), and oesophagus (Harms 2005) showing that treatment effects with respect to tumour control and toxicities are comparable with continuous LDR as long as fractionation rules are followed. However in a series of patients treated for anal canal cancer with a PDR boost, a high tumour control (88%) but a higher than ex- pected complication rate in the form of local necrosis and ulcer- ation were noted in 13 out of 17 patients (Roed 1996). A colosto- my was required in eight. This was attributed to an improper BT technique, not following the rules of the Paris System. Some authors introduced daytime PDR schedules aiming for an ambulant treatment and a reduced overall cost (Brenner 1997). This “office hours” schedule is not supported by any study re- porting long-term results. Probably, only the complete 24–hour treatment schedule preserves the radiobiological advantages of LDR BT. 9.5 Low dose rate BT versus high dose rate BT There have been some historical comparisons (Fu 1990, Or- ton1991).and a few randomised trials comparing LDR and HDR BT in cervix cancer and oral tongue cancer (Inoue 1996, In- oue1998, Patel1994, Shigematsu 1983). However, no trial met the criteria of modern randomised studies. In cervix cancer, HDR as compared with LDR BT produced similar local control rates but fewer complications. This is explained by the advantages of mod- ern gynaecological applicators with fixed geometry and stepping source technology for adjusting dose distribution, as compared with old fashioned radium sources, without any fixed geometry or optimisation possibilities rather than by a dose rate effect (Or- ton1991 Patel1994, Shigematsu 1983). In some specific circumstances, LDR BT might nevertheless be less toxic for late responding normal tissues, for instance when it is necessary to reach the limit of tolerance to maximise local con- trol rate. Examples are exclusive BT for small cancers of the oral cavity (Lau 1996) or treatment with irradiation (or chemoradi- ation) in locally advanced cancer of the cervix (Petereit 1999).