In Vivo Dosimetry using Plastic Scintillation Detectors for External Beam Radiation Therapy

2.4 In Vivo Dosimetry 
The dosimetric properties of PSDs make many useful applications possible such as small 
field dosimetry (Beddar et al. 2001, Klein et al. 2010), quality assurance (Gagnon et al. 
2012), and in vivo dosimetry. Among these applications, in vivo dosimetry has attracted 
great interest. The benefits of in vivo dosimetry will be laid out in the rest of this section. 
In vivo dosimetry is of interest primarily for its potential to improve patient safety 
and verify correct delivery of treatment. Patient safety is imperative in radiation therapy. 
Patients are exposed to high levels of radiation and both over- and under-exposure can 
have severe consequences as illustrated in figure 2.4. These consequences have been 
highlighted by recently reported accidents. In Panama 28 patients received excessive dose 
during treatment between August 2000 and March 2001 due to an error in the way the 
treatment planning system digitized shielding blocks. Eight of the patients subsequently 
died, with five deaths attributed to the overdose. The remaining patients were expected to 
develop complications (IAEA 2001). In Glasgow in 2006, human error induced by a 
change in the way dose was specified resulted in a medulloblastoma patient receiving 55 
Gy in 19 fractions instead of the intended 35 Gy in 20 fractions. This overdose eventually 
18 

---

Figure 2.4. Hypothetical tumor cure probability (TCP) curve and normal tissue 
complication probability (NTCP) are plotted in black. Successful radiation therapy 
maximizes the difference between the tumor cure probability and normal tissue 
complication probability to achieve the highest likelihood of uncomplicated cure (green 
dashed curve). The likelihood of tumor recurrence is plotted in blue to illustrate that 
deviation from the optimal dose in either direction can significantly increase the 
likelihood of either recurrence (caused by under-dose) or healthy tissue damage (caused 
by over-dose). 
19 

---

resulted in the patient’s death (Mayles 2007). Some errors are less severe but affect far 
more patients. For example, at a French center between 2001 and 2006, 397 patients 
received 8% overdoses because the dose resulting from MV portal imaging was not 
included in planning. No patients died, but the population exhibited an abnormally high 
rate of radiation induced complications (Derreumaux et al. 2008). In each of these cases, 
in vivo dosimetry could have detected errors early in the course of treatment, sparing 
patients undue injury through timely corrective action.
In vivo dosimetry also has the potential to improve the patient experience. The 
promulgation of reports of radiation therapy accidents by the media produce anxiety in 
some patients undergoing treatment (The Royal College of Radiologists 2008). In vivo 
dosimetry can reassure patients that errors will be caught and mitigated if they occur, and 
bolster the patient’s confidence in the clinic. Thus, the perceived quality of care is 
improved even when no deviations in treatment occur. 
Another motivation to adopt in vivo dosimetry is that it may be legally required in 
the future. Some European governments have begun mandating in vivo dosimetry in 
response to accidents similar to those cited above. It is required by law in France, 
Sweden, and Denmark. The National Health Service in Britain recommended in 2008 that 
routine in vivo dosimetry be implemented for all patients undergoing radiation therapy. 
While in vivo dosimetry is not required on a routine basis in America, it is reasonable to 
assume regulation could move in that direction in the future. 
Finally, in vivo dosimetry is of scientific interest as well, because it generates data 
useful for toxicity studies. Tumor control probabilities and normal tissue risks are 
evaluated by correlating outcomes with planned doses as calculated by a treatment 
20 

---

planning system (TPS). While TPSs generally do an excellent job of accurately 
calculating dose distributions, the TPS calculated dose cannot account for day to day 
variations in setup and other external factors that affect the delivered dose distribution. In 
vivo dosimetry can be used to evaluate such effects, and may therefore be useful when 
used in conjunction with TPS calculated dose for evaluating toxicity risks.