Radiotherapy Interruption in Cancer Patients: Rates and Causes at Ahvaz Golestan Hospital

Cancer is one of the most common causes of death all over the world which is spreading fast. Different types of treatment modalities are used in different types of cancers; such as surgery, chemotherapy, and radiotherapy [1]. Radiotherapy is based on exposing malignant cells to ionizing radiation, which can lead to death of these cells [2]. The ideal goal of radiotherapy is the delivery of more doses to tumor cells and less doses to the surrounding normal tissues to produce a high probability of tumor control whilst causing minimal normal tissue complications [3]. A conventional course of radiotherapy is scheduled for five days per week over several weeks; however, interruptions in this course may occur and unplanned gaps are a common occurrence. These gaps and interruptions in Abstract

Introduction treatment regime may lead to the repopulation of tumor clonogens [4]. In particular for the head and neck cancers for which the evidence is the strongest for accelerated repopulation of clonogenic tumor cells, it is estimated that each one and seven days interruptions in treatment course result in 0.68-1.4% and 14-20% reduction in the local control rate, respectively [5-6-7].
Completion of radiotherapy schedule has special importance for increasing tumor control probability and reducing the probability of tumor progression and/ or recurrence [8-9-10]. Thus, one of the main goals in radiation therapy is the delivery of the total dose without any interruptions or prolongation of the overall treatment time [5-9-11].
Based on above discussion, the interruption in 1 Assistant Professor of Radiotherapy and Oncology, Department of Clinical Oncology, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran. 2

Materials and Methods
Cancer patients treated at the Ahvaz Golestan Hospital during March 2012 to March 2013 were included in this retrospective analytical-descriptive study which was done at 2015. All patients treated with Varian Clinac 2100C or Siemens Primus linear accelerators. Patients were included whether their radiation treatment was curative or palliative. Patients treated with no treatment gap were excluded from the study. Interruptions in treatment were considered as the number of calendar days over which it would have been possible to have treatment (i.e., weekdays). Weekends were not regarded as missed appointments. The required information in the questionnaire included age, gender, site of cancer (head and neck v/s other types), intent of treatment (curative v/s palliative), interruption period, causes of interruption, patients' address and phone numbers.
The Statistical Package for the Social Sciences (SPSS) version 15 was used to analyze the obtained data. The descriptive statistics were average and standard deviation in quantitive variables, and frequency and percent in qualitative variables. In order to found any relation between different variables, the Fischer, independent T, Chi score and regression logistic tests were applied. P-values of < 0.05 were considered significant.

Results
Among 1476 patients of this study, 861 were female (58.3%) and 615 were male (41.7%). About 32.7% of male patients (201 from 861 patients) and 26.8% of female patients (231 from 615 patients) had interruption in their radiotherapy course (Table 1). There was no significant correlation between patients' gender and interruption probability (P-value: 0.055, OR: 1.254).
The patients' ages were in the range of 1 to 96 years old with the average age of 51.72 years with 18.004 standard deviation. In 1044 patients without interruption and 432 interrupted patients, average ages were 51.87 and 51.22 years old, respectively. There was no significant relation between patients' ages and interruption probability (P-value: 0.535).
Based on the cancer site, the patients were divided in two categories; 376 patients (25.5%) had head and neck cancer and the remaindered 1100 patients (74.5%) had cancers of other sites. More than 35.1% of 376 patients with head and neck cancer (132 patients) and 27.3% of 1100 patients with other sites of cancer (300 patients) had interruption in their treatment courses. The probability of treatment interruption in head and neck cancers was significantly higher than other cancers (P-value: 0.014, OR: 1.377).
Also, the patients were categorized for their living place as follows: 568 patients (38.5%) were in Ahvaz city, 738 patients (50%) in Khuzestan province but not in Ahvaz city, and the remaindered 170 patients (11.5%) out of Khuzestan province. About 30.3% of 568 Ahvazi patients (172 patients), 26.5% of 738 non-Ahvazi patients from Khuzestan province (196 patients), and 37.9% of 170 patients from out of Khuzestan province (64 patients) had interruption in their treatment courses. There were no statistically significant differences in interruption probability among Ahvazi and non-Ahvazi patients from Khuzestan province (P-value: 0.134). However; the probability of treatment interruption in patients from out of Khuzestan province was significantly higher than Ahvazi (P-value: 0.093, OR: 1.362) and non-Ahvazi patients from    [12] showed that 76.6% of 478 cancer patients with all primary sites treated with curative intent had radiotherapy interruption and 17.9% of the interruptions were greater than 5 days and 5.6% greater than 10 days. There are some differences in the causes of interruption in Ahvaz and those reported in other centers. Public holidays are one of the most common causes of treatment gaps; however in Ahvaz holidays accounted for 8.1% of treatment interruptions, whereas in the United Kingdom they were responsible for 39%, in the Spanish study 45%, and in Vancouver for 69% of treatment interruptions. This is because of efforts aimed at working our staff and machines in the most public holidays in Ahvaz.
Machine breakdown and maintenance account for 44% of treatment interruptions in the United Kingdom, 45% in Spain, and 2% in Vancouver. About 29.4% of gaps were due to this reason in Ahvaz. This is because of efforts aimed at treating patients with another machine, despite of high rate of machine break down at our center.
The second most common cause of interruption in Ahvaz was treatment side effect (16.7%) which was higher than other centers and equal to older studies like the study of Barton et al in which radiation side effects was responsible for 16% of all treatment breaks. This rate was 2% in Spain, 10% in Vancouver, and 8% in the United Kingdom. This is because of lack of a multidisciplinary team to support patients to reduce treatments missed from acute toxicities.
In conclusions, unplanned radiotherapy interruption is a major problem in normal clinical practice. Generally, treatment interruption at our center was not much higher than other centers; however, the most common cause of treatment interruption was equipment damages and/or maintenance, which was higher than most centers. This important issue requires special attention of authorities. Most interruptions are short and can be planned by treating patients on weekends and holidays to maintain the OTT.
Moreover, based on the intent of treatment, the patients were divided into two groups; 1302 patients had curative treatment (88.2%) and 174 patients had palliative treatment (11.7%).
There was no any interruption in 70.7% of patients (1044 people) in the treatment course (Table 1 and 2). On the other hand, the treatment interruption was found in 29.3% of patients (432 people), 13.2% of patients (195 people) with 1 to 3 days interruption period, 11.2% of patients (165 people) with 4 to 30 days interruption period, and 4.9% of patients (72 people) with more than 30 days interruption period ( Table 3).
The most common cause of interruption in the treatment course was equipment damages and/or maintenance with 29.5% (127 patients) of all causes. Other causes of treatment interruption were treatment side effects and general conditions (16.7%), public holidays (8.1%), patients' death (5.3%), trip difficulties (4.7%), treatment cost (1.9%), and miscellaneous causes (shifting to another center and personal problems or tendencies in 7.9%). In 112 patients (25.8%) no clear reason was found and the patients were not accessable for more investigation ( Table 2).

Discussion
More than 35.1% of patients with head and neck cancer and 27.3% of patients with other sites of cancer had interruption in their treatment courses, which this difference was statistically significant (P-value: 0.014, OR: 1.377). This is in contrast with a Spanish study by Garau et al in 2008 in which only 23.4% of patients finished their radiotherapy in the planned overall treatment time (OTT) but 48.9% of head and neck cancer patients finished their treatment in the planned OTT. One of the main reasons for this difference is that for the head and neck cancers, the patients receive more doses of radiation due to curative and radical intent of treatment in these patients, hence showing more severe and debilitating treatment side effects. Also in general, these patients have low socioeconomic status with less social and family support leading to increasing rates of interruption. The Royal College of Radiographers [12][13] recommends that proactive support from a multidisciplinary team of healthcare professionals and counselors be available to patients to reduce treatments missed from acute toxicities. Unfortunately we don't have this team at our center.
About 30.3% of Ahvazi patients, 26.5% of non-Ahvazi patients from Khuzestan province, and 37.9% of patients from out of Khuzestan province had interruption in their treatment courses. The probability of treatment interruption in patients from out of Khuzestan province was significantly higher than Ahvazi (P-value: 0.093, OR: 1.362) and non-Ahvazi patients from Khuzestan province (P-value: 0.006, OR: 1.642). This difference is most likely due to the distance of living place to the radiotherapy center.
The treatment interruption was found in 29.3% of patients, 13.2% of patients with 1 to 3 days gap, 11.2% of