

ORIGINAL ARTICLE 

Ahead of print publication 


Radiobiological modeling of radiationinduced acute rectal mucositis: A singleinstitutional study of cervical carcinoma
Balbir Singh^{1}, Gaganpreet Singh^{2}, Arun Singh Oinam^{3}, Vivek Kumar^{4}, Rajesh Vashistha^{5}, Manjinder Singh Sidhu^{5}, Maninder Singh^{5}
^{1} Centre for Medical Physics, Panjab University, Chandigarh; Department of Radiation Oncology, Max Superspeciality Hospital, Bathinda, Punjab, India ^{2} Centre for Medical Physics, Panjab University; Department of Radiotherapy, PGIMER, Chandigarh, India ^{3} Department of Radiotherapy, PGIMER, Chandigarh, India ^{4} Centre for Medical Physics, Panjab University, Chandigarh, India ^{5} Department of Radiation Oncology, Max Superspeciality Hospital, Bathinda, Punjab, India
Date of Submission  01Jun2021 
Date of Acceptance  02Jul2021 
Date of Web Publication  09Oct2021 
Correspondence Address: Arun Singh Oinam, Department of Radiotherapy, PGIMER, Chandigarh  160 012, India
Source of Support: None, Conflict of Interest: None DOI: 10.4103/jcrt.jcrt_879_21
Purpose: This study aimed to estimate the fitting parameters of sigmoidal dose–response (SDR) curve of radiationinduced acute rectal mucositis in pelvic cancer patients treated with Intensity Modulated Radiation Therapy (IMRT) for the calculation of normal tissue complication probability (NTCP). Materials and Methods: Thirty cervical cancer patients were enrolled to model the SDR curve for rectal mucositis. The patients were evaluated weekly for acute radiationinduced (ARI) rectal mucositis toxicity and their scoring was performed as per Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. The radiobiological parameters, namely n, m, TD_{50}, and γ_{50} were calculated from the fitted SDR curve obtained from the clinical data of cervical cancer patients. Results: ARI toxicity for rectal mucosa in carcinoma of cervical cancer patients was calculated for the endpoint rectal mucositis. The n, m, TD_{50}, and γ_{50} parameters from the SDR curve of Grade 1 and Grade 2 rectal mucositis were found to be 0.328, 0.047, 25.44 ± 1.21 (confidence interval [CI]: 95%), and 8.36 and 0.13, 0.07, 38.06 ± 2.94 (CI: 95%), and 5.15, respectively. Conclusion: This study presents the fitting parameters for NTCP calculation of Grade 1 and Grade 2 ARI rectal toxicity for the endpoint of rectal mucositis. The provided nomograms of volume versus complication and dose versus complication for different grades of rectal mucositis help radiation oncologists to decide the limiting dose to reduce the acute toxicities.
Keywords: Acute radiation toxicity, CTCAE, normal tissue complication probability, rectal mucositis
How to cite this URL: Singh B, Singh G, Oinam AS, Kumar V, Vashistha R, Sidhu MS, Singh M. Radiobiological modeling of radiationinduced acute rectal mucositis: A singleinstitutional study of cervical carcinoma. J Can Res Ther [Epub ahead of print] [cited 2022 Dec 6]. Available from: https://www.cancerjournal.net/preprintarticle.asp?id=327878 
> Introduction   
Cervical cancer is the second most common malignancy in incidence and the second most common cause of cancerrelated mortality amongst women.^{[1]} Most of these patients are presented in locally advanced stages for which definitive radiotherapy with concurrent chemotherapy is the choice of treatment. It is because the combination of these two modalities improves tumor control, overall survival, and progressionfree survival.^{[2]} Radiotherapy for cervical cancer can lead to several gastrointestinal complications such as proctitis, cystitis, and anorectal dysfunction.^{[3]} However, unlike small bowel and colon, the effect of radiation in rectal stem cells has not been explored extensively. Radiationinduced rectal epithelial damage is a widespread side effect of pelvic radiotherapy, which often leads to pain and weight loss. Hence, compromising the quality of life (QOL) may also lead to treatment interruptions resulting in poor treatment outcomes.^{[4]} Acute radiationinduced rectal mucositis usually manifests itself during or shortly after irradiation.^{} Computation of radiobiological parameters for radiationinduced acute rectal mucosal toxicity in radiotherapy treatment planning plays an important role in the estimation of normal tissue complication probability (NTCP). The incidence of acute toxicity is common between external beam radiotherapy (EBRT) and chemotherapy.^{[2]} The effects of both treatment modalities are synergistic. Based on multiple established studies and clinical experience of radiation oncologists, normal or healthy tissues present close to the tumor receive the high doses as the target volume, thus producing the normal tissue toxicity.^{[3],[4],[5]} Most of the radiobiological models utilize the dosimetric and volumetric data in terms of dose–volume histograms (DVHs) of the irradiated organs and tumor volumes in the treatment plan to calculate the NTCP and tumor control probability.^{[6],[7],[8]} Rubin and Cassarett initially published normal tissues and organs tolerance doses in the form of TD_{5/5} (5% complication in 5 years) and TD_{50/5} (50% complication in 5 years) for calculation of NTCP.^{[9]} In the study of Emami et al., normal tissue tolerance doses for only few selected normal organs were calculated as a function of whole and partial organ volumes in the same form of TD_{5/5} and TD_{50/5}. In these studies, doses were calculated for uniformly irradiated onethird, twothird, and the whole volume of organs by using a standard fractionation scheme, i.e. 1.8–2.0 Gy per fraction.^{[10]} Many radiobiological models have compiled dose tolerance data for several normal tissues/organs, but there is still a lack of clinical data for all of the tissues/organs. Moreover, radiobiological endpoints calculated in these models are derived from the clinical data related to the late toxicities of the tissues.^{[11],[12],[13],[14]} In this study, acute rectal mucositis is considered because there is a lack of availability of data for Grade 1 and Grade 2. Acute tissue toxicity in terms of radiobiological endpoints for Grade 1 and Grade 2 toxicity of rectal mucosa was investigated for the pelvic cancer patients treated with intensitymodulated radiotherapy therapy (IMRT) technique. The mathematical modeling of the dose–response relationship curve requires a number of parameters that can be utilized to define the dose constraints of the organ of interest for safe and welltolerated doses.^{[12]} This study aims to calculate the fitting parameters corresponding to the different radiobiological endpoint of acute radiation toxicity in pelvic cancer patients during EBRT and utilize these derived parameters to determine the NTCP under the condition of inhomogeneous radiation.
> Materials And Methods   
Patient selection
Thirty patients with pelvic cancer were selected for this study, and their characteristics are shown in [Table 1]. The selected patients have a range of pelvic primary toxicity sites including endometrial, cervix, and ovary sites.  Table 1: Patient (n=30) and tumor characteristics in patients with pelvic tumors treated via intensitymodulated radiation therapy
Click here to view 
Patient simulation and delineation
A planning computed tomography (CT) scan of each patient was taken in a supine position using 16 multislice CT scanner (Light Speed 16 Gold Seal, Wipro GE Healthcare Technologies, WI, USA) with a slice thickness of 5 mm. Patients were immobilized with four clamps thermoplastic cast. All the patients were scanned from the upper border of liver to the mid shaft of femur to include the primary site. The delineation of organs at risks (OARs), gross tumor volume, nodes (if any), and clinical target volumes was performed on FOCAL SIM Version 5.11, (Elekta CMS, Maryland Heights, MO, USA) using Radiation Therapy Oncology Group guidelines of pelvic cancer, as shown in [Figure 1].^{[15]} The rectal mucosa was delineated from rectosigmoid junction to pubic symphysis. All CT images, along with the delineated structure sets, were exported to the Computerized Medical System CMS XiO version 5.10 (Elekta, Stockholm AB, Sweden) treatment planning system (TPS).  Figure 1: Delineated organs at risk: rectal mucosa (red color) and bladder (violet color) over the axial (top left), coronal (bottom left), sagittal (bottom right), and 3D view (top right)
Click here to view 
Treatment planning
CMS XiO (version 5.10) TPS (Elekta, Stockholm AB, Sweden) was used to create the simultaneous integrated boost IMRT radiotherapy treatment plans of the pelvic patients. Two prescription dose regimens were chosen, i.e. 45 Gy and 50 Gy corresponding to the nodes and planning target volume (PTV) in 25 fractions, respectively, and 45 Gy and 50.4 Gy corresponding to the nodes and PTV in 28 fractions, respectively, as shown in [Table 2]. The mean coverage of PTV and nodes by 95% of the prescribed dose was chosen as a plan evaluation parameter according to the International Commission on Radiation Units and Measurements (ICRU) 83.^{[16]} All IMRT plans were made with nine fields of 6 megavoltage Xrays beam with step and shoot delivery technique using superposition algorithm having segmentation method of smart sequencing and minimum segment area of 2 cm^{2} for Elekta (Synergy Platform) linear accelerator equipped with 40 pairs of multileaf collimators, and grid size of 2 mm was chosen for the dose calculation. To verify the IMRT treatment plan's dose delivery, a pretreatment patientspecific quality assurance (passing/failure criteria) procedure was performed for each patient with OmniPro IMRT verification software version 1.77.0021 using a 2D array of ion chamber I'mRT MatriXX (Scanditronix Wellhofer, Freiburg, Germany) which consists of 1020 ion chambers. The gamma evaluation criterion was chosen as 3 mm for dose to agreement and 3% for dose difference.
Assessment of acute radiationinduced toxicity: Rectal mucositis
Radiationinduced acute toxicity was assessed for the radiobiological endpoint, i.e. rectal mucositis. Mucosal reactions of the rectal mucosa were clinically assessed by welltrained physicians/radiation oncologists. Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 were used to grade acute radiation toxicities in each patient, and the entire grades are classified into five categories for toxicity assessment, as shown in [Table 3].^{[17]} This toxicity was documented once per week during radiation and 2 weeks after the end of the radiation therapy treatment for a total duration of 6 weeks.  Table 3: Common terminology criteria adverse events for scoring toxicity
Click here to view 
Mathematical modeling
The clinical outcome of the patients having different grades of toxicity was correlated with the delivered dose during the course of radiation therapy treatment. The sigmoidal dose–response (SDR) curve fitting was performed using the data of (1) mean and partial volume doses of OARs calculated from the DVHs, (2) the number of fractions delivered, and (3) time of onset of the acute radiation toxicities which were taken as input parameters. The radiobiological parameters n, m, TD_{50} and γ_{50} calculated from the SDR were also utilized to calculate the NTCP of rectal mucosa.
> Results   
All 30 patients who underwent radiotherapy course of treatment were clinically evaluated for the acute toxicities (during the treatment and also on a weekly basis). Their responses were recorded as per the CTCAE criteria. The number of patients who appeared with different grades of acute toxicity is shown in [Table 4].
Radiobiological modeling and its parameters
The change of complication probability with change in dose and volume is plotted in the surface plot form. [Figure 2] and [Figure 3] show the dose–response curve of Grade 1 and Grade 2 of rectal mucositis. In each of these figures, (1) Figure (a) represents the nomogram of NTCP and volume of the partially irradiated OARs for three different tolerance doses corresponding to the 5%, 20%, and 50% complication; (2) Figure (b) shows the complication probability as a function of dose for onethird, twothird, and whole volume of the OARs and shows threshold dose behavior for the NTCP (3), Figure (c) represents the change of constant NTCP corresponding to the change of volume of irradiation of the OARs as a function of dose, and (4) Figure (d) represents the surface plot of tolerance dose on one axis, partial volume on the second axis against the complication probability on the third axis.  Figure 2: Rectal Mucositis Grade 1 nomogram, (a) Nomogram of complication probability as a function of partial volume for partially irradiated OAR for three different tolerance doses (b) Nomogram of complication probability as a function of dose for 1/3rd, 2/3rd and whole volume of the OAR (c)Nomogram of partial volume as a function of dose for 5%, 20% and 50% complication (d) Surface plot of tolerance dose on one axis, partial volume on second axis against the complication probability on the third axis
Click here to view 
 Figure 3: Rectal mucositis Grade 2 nomogram, (a) Nomogram of complication probability as a function of partial volume for partially irradiated OAR for three different tolerance doses (b) Nomogram of complication probability as a function of dose for 1/3rd, 2/3rd and whole volume of the OAR (c) Nomogram of partial volume as a function of dose for 5%, 20% and 50% complication (d) Surface plot of tolerance dose on one axis, partial volume on second axis against the complication probability on the third axis
Click here to view 
The values of n, m, TD_{50}, and γ_{50} parameters for Grade 1 and Grade 2 rectal mucositis are calculated from the fitted SDR curve, as shown in [Table 5].  Table 5: Radiobiological parameters obtained from the sigmoidal doseresponse curve fitting
Click here to view 
> Discussion   
It is a wellknown fact that radiation therapy causes an impact on QOL.^{[18]} The association between the doses received by the rectal mucosa and incidence of patientreported acute rectal mucosal toxicity in the pelvic cancer patient population is still not known for Grade 1 and Grade 2 toxicity. This study has addressed this critical knowledge gap. The incidence of Grade 3 and Grade 4 of acute rectal mucositis toxicity (AMT) was found to be low despite the usage of high radiation doses in cervical cancer patients. This is because of the usage of conformal radiation technique, i.e., IMRT.
In the study of Burman et al., dose–volume parameters based on normal tissue tolerance doses were given only for a few number of organs for different radiobiological endpoints and all these radiobiological endpoints were calculated based on the 2D radiation delivery technique for late tissue toxicity.^{[14]} However, these dose–volume parameters and their associated normal tissue toxicity data are revised with the availability of recent clinical data and advanced radiotherapy techniques like conformal radiotherapy, but there is nonavailability of radiobiological parameters for many normal tissues which can estimate NTCP.^{[19]} The rectal mucosa is one of them and their AMT have been studied in the present study for the endpoint of rectal mucositis. The volume of 2 cc, 6 cc, and 20 cc of rectal mucosa was used to calculate the reference area of 10, 30, and 100 cm^{2}, respectively, by dividing the volume with the 0.2 cm thickness of rectal mucosa. The tolerance doses of rectal mucosal toxicity were calculated for the reference area of 10, 30, and 100 cm^{2} of these organs, respectively, and are similar to the methodology used in Emami et al., as tolerance dose was calculated for skin.^{[14]}
Emami et al. initially made an attempt for determining the complication probability of late rectum toxicity.^{[10]} This model can predict the late rectal mucosal toxicity only for the endpoint of proctitis for Grade 3 or above but not for rectal mucositis as both the endpoints are different given in CTCAE V5. All the radiobiological parameters determined in this study were only for late radiationinduced effects. However, no radiobiological parameter and SDR curve were calculated for acute effects of radiation which can predict Grade 1, Grade 2, and above rectal mucosal toxicity.
Michalski et al. had studied rectal toxicity based on dose–volume effect of the rectum.^{[20]} The volume receiving more than 60 Gy can produce Grade 2 or more toxicity corresponding to the endpoint of rectal bleeding. To determine these toxicity data, Lyman–Kutcher–Burman model normal tissue complication parameters were used. There are no radiobiological parameters determined in his study which can predict the Grade 1 and Grade 2 rectum toxicity.
In the present study, various radiobiological fitting parameters from the SDR curve of rectal mucosa were determined. The parameter “n” value is specifically unique for each organ which describes the volumetric dependence of the dose–response relationship. The complication probability curves for a tissue depend upon parameter “n.” The volume dependency for the occurrence of complication rises as the value of “n” increases. The value of “n” is 0.32 and 0.13 for G1 and G2 rectal mucosa, respectively, which is more than zero for both the normal tissues, as shown in [Figure 2]d and [Figure 3]d. Therefore, complication probability produced has small volume dependence of the tissue being irradiated. These curves are shown in [Figure 2]a and [Figure 3]a are for G1 and G2 rectal mucositis and representing complication probability as a function of partial volume keeping the dose constant. The SDR curves of rectal mucositis for both grades show threshold type behavior of complication probability for any given dose. The complication probability is not varying with partial volume until a certain volume gets irradiated. At low partial volumes, the toxicity rises faster; as the dose increases, complication probability increases. As the TD_{50} values increase, the severity of complication also increases gradually for the radiobiological endpoint for different toxicity grades.
The tolerance doses for 5%, 20%, and 50% complication of the irradiated tissues were calculated for whole, twothird, and onethird uniform organ irradiation as shown in [Figure 2]b and [Figure 3]b. For any constant partial volume of rectum, complication probability rises as the dose is increased.
[Figure 2]c and [Figure 3]c represent the partial volume as a function of dose, keeping the complication probability constant for rectal mucositis for both the grades corresponding to complication probability of 5%, 20%, and 50%. As the dose is increased for constant complication, the irradiated partial volume is decreased.
Any dose–volumerelated complication probabilities can also be calculated using the nomogram diagram showing in [Figure 2]d and [Figure 3]d.
The SDR curve for normal tissue is the probability of causing complications in an organ as a function of dose.^{[21]} The dose at which 50% response probability is achieved is TD_{50} and γ_{50} is the slope normalized at 50% complication for a given dose. The parameter m determines the slope of the curve for complication probability as a function of dose.^{[14]} The slight variation in m and TD_{50} changes the complication probability, as shown in [Figure 2]b and [Figure 3]b, representing the complication probability as a function of dose keeping the partial volume constant for whole, twothird, and onethird volume of irradiation. As the dose is increased, complication probability is also increased for any fixed irradiated partial volume.
The results of radiobiological parameters which are extracted from SDR curves can be used to estimate NTCP of rectal mucosa in cervical cancer patients.
> Conclusion   
In this study, radiobiological parameters were derived successfully from the SDR curve for Grade 1 and Grade 2 acute rectal mucosal toxicity of individual pelvic cancer patients. These parameters can be used to estimate the dependence of toxicity in this organ, which can predict the acute rectal mucositis in a large patient cohort. Using nomogram of Grade 1 and Grade 2 toxicity, any reader can manually find the dose–volumerelated complication probabilities of acute rectal mucositis. Late adverse effects in the rectal mucosa can also be examined in some other studies. After calculating the radiobiological parameters, it increases the clinicians' confidence in taking any clinical decisions during plan evaluation. In addition, our results need to be validated in a larger cohort of patients.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
> References   
1.  Petignat P, Roy M. Diagnosis and management of cervical cancer. Bmj 2007;335:765–8. 
2.  Radojevic MZ, Tomasevic A, Karapandzic VP, Milosavljevic N, Jankovic S, Folic M. Acute chemoradiotherapy toxicity in cervical cancer patients. Open Med (Wars) 2020;15:82232. 
3.  Rubin P, Cassarett GW. Urinary tract: The kidney. Clin Radiat Pathol 1968;1:293333. 
4.  Tirado FR, Bhanja P, CastroNallar E, Olea XD, Salamanca C, Saha S. Radiationinduced toxicity in rectal epithelial stem cell contributes to acute radiation injury in rectum. Stem Cell Res Ther 2021;12:1–14. 
5.  Warkentin B. Radiobiological modelling in radiation oncology. Med Phys 2008;35:1621. 
6.  Mesbahi A, Rasouli N, Mohammadzadeh M, Nasiri MB, Ozan TH. Comparison of radiobiological models for radiation therapy plans of prostate cancer: Threedimensional conformal versus intensity modulated radiation therapy. J Biomed Phys Eng 2019;9:26778. 
7.  Oinam AS, Singh L, Shukla A, Ghoshal S, Kapoor R, Sharma SC. Dose volume histogram analysis and comparison of different radiobiological models using inhouse developed software. J Med Phys 2011;36:2209. [ PUBMED] [Full text] 
8.  Kehwar TS. Analytical approach to estimate normal tissue complication probability using best fit of normal tissue tolerance doses into the NTCP equation of the linear quadratic model. J Cancer Res Ther 2005;1:16879. 
9.  Rubin P, Casarett G. Direction for clinical radiation pathology. The tolerance dose. Front Radiat Ther Oncol 1972;6:116. 
10.  Emami B. Tolerance of Normal Tissue to Irradiation. Int J Radiat Oncol Biol Phys 1991;21:10922. 
11.  Coia L, Emami B, Solin LJ, Munzenrider JE, Lyman J, Shank B, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991;21:10922. 
12.  Mavroidis P, Pearlstein KA, Dooley J, Sun J, Saripalli S, Das SK, et al. Fitting NTCP models to bladder doses and acute urinary symptoms during postprostatectomy radiotherapy. Radiat Oncol 2018;13:17. 
13.  Niemierko A, Goitein M. Modeling of normal tissue response to radiation: The critical volume model. Int J Radiat Oncol Biol Phys 1993;25:13545. 
14.  Burman C, Kutcher GJ, Emami B, Goitein M. Fitting of normal tissue tolerance data to an analytic function. Int J Radiat Oncol Biol Phys 1991;21:12335. 
15.  Nitsche M, Brannath W, Brückner M, Wagner D, Kaltenborn A, Temme N, et al. Comparison of different contouring definitions of the rectum as organ at risk (OAR) and dosevolume parameters predicting rectal inflammation in radiotherapy of prostate cancer: Which definition to use? Br J Radiol 2017;90:112. 
16.  Gregoire V, Mackie TR. Dose prescription, reporting and recording in intensitymodulated radiation therapy: a digest of the ICRU Report 83. Imaging Med 2011;3:367 
17.  
18.  Liu L, Meers K, Capurso A, Engebretson TO, Glicksman AS. The Impact of Radiation Therapy on Quality of Life in Patients withCancer. Cancer Pract 1998;6:23742. 
19.  Bentzen SM, Constine LS, Deasy JO, Eisbruch A, Jackson A, Marks LB, et al. Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC): an introduction to the scientific issues. Int J Radiat Oncol Biol Phys 2010;76:S39. 
20.  Michalski JM, Gay H, Jackson A, Tucker SL, Deasy JO. Radiation dosevolume effects in radiat1. Michalski JM, Gay H, Jackson A, Tucker SL, Deasy JO. Radiation dosevolume effects in radiationinduced rectal injury. Int J Radiat Oncol Biol Phys. 2010;76(3):S123S129. ioninduced rectal injury. Int J Radiat Oncol Biol Phys 2010;76:S1239. 
21.  Stavreva N, Stavrev P, Warkentin B, Fallone BG. Derivation of the expressions for $γ$50 and D50 for different individual TCP and NTCP models. Phys Med Biol 2002;47:3591. 
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]
