Volumetric and dosimetric comparison of organs at risk between the prone and supine positions in postoperative radiotherapy for prostate cancer

It is widely known that adjuvant radiotherapy (RT) after prostatectomy for prostate cancer with adverse pathological findings improves the biochemical progression-free survival, local control, metastasis-free survival, and overall survival [1‐5]. In cases in which a patient shows prostate-specific antigen (PSA) relapse after radical prostatectomy, salvage RT also decreases the risk of local recurrence and metastasis and improves prostate cancer-specific survival [6, 7]. Dose escalation contributes to a good prognosis in patients treated with postoperative RT after prostatectomy [8‐10]. Although intensity modulated radiation therapy (IMRT) enables us to reduce the risk of radiation-induced toxicity [11, 12], dose escalation is limited by the organs at risk (OARs) surrounding the planning target volume (PTV), particularly the small intestine.

According to a report on uterine cervical cancer, high-dose irradiation is a risk factor for perforation of the small intestine [13]. When treating the pelvic area with RT, it is sometimes necessary to reduce the PTV because the small intestine is located in or near the PTV. A systematic review on the use of an absorbable hydrogel spacer revealed that the placement of a spacer between the prostate and rectum reduced late rectal toxicity and improved the bowel, urinary, and sexual quality of life (QOL) in patients undergoing IMRT for prostate cancer [14]. From this result, the space between the small intestine and the PTV is also assumed to reduce toxicity not only in the small intestine, but in other OARs by making it easier to observe the dose constraints of the other OARs.

Patient positioning may affect the positional relationship among the PTV and the OARs. The patient is sometimes placed in the prone position for RT for pelvic malignancies to reduce the fraction of the small intestine that is exposed to a high dose of radiation. Some studies of three-dimensional conformal radiotherapy (3D-CRT) or IMRT for pelvic malignancies showed that the dose to the small intestine in the prone position was lower than that in the supine position [15, 16], whereas others showed that the dose to the small intestine did not differ to a statistically significant extent between the two patient positions [17, 18]. There have so far been almost no studies about RT in the prone position for prostate cancer in the postoperative setting.

The aim of the present study was to evaluate the effect of patient positioning on the volume of the OARs in or near the PTV and on the dose of the OARs in adjuvant or salvage RT for prostate cancer after prostatectomy.

Patients received planning CT scans from March 2017 through November 2017. The patient characteristics are summarized in Table 2. The median age was 71 years (range, 67–84). The median (BMI) was 23.5 (20.3–26.7). The median time between prostatectomy and the initiation of RT was 24.9 (2–188) months. Seven patients (41.2% of the patients) received adjuvant RT and ten patients (58.8%) received salvage RT.

In four patients (23.5%), the small intestine overlapped with the PTV in the supine position, but no part of the small intestine was included in the PTV in the prone position. Although the small intestine and the PTV overlapped in both the prone and the supine position in two patients (11.8%), the overlapping volumes were reduced in the prone position in both patients. Dosimetric comparison was conducted in 11 patients (64.7%), whose small intestine was not within the PTV in both the prone and supine positions. All 17 patients were included in volumetric analysis. The mean doses of the periphery of the PTV, PTV + 0.5 cm, PTV + 1 cm, PTV + 2 cm, PTV + 3 cm, and PTV + 4 cm were 63 Gy, 48 Gy, 34 Gy, 23Gy, 16 Gy, and 12 Gy, respectively.

In volumetric analysis in the 17 patients, the overlapping volumes of the OARs are shown in Table 3. The overlapping volumes of the small intestine were as follows (prone position [mean ± SD] vs. supine position, respectively): PTV, 1.5 ± 5.5 cm³ vs. 7.9 ± 15.7 cm³ (P = 0.036); PTV + 0.5 cm, 2.6 ± 8.9 cm³ vs. 12.1 ± 22.6 cm³ (P = 0.035); PTV + 1 cm, 3.5 ± 11.4 cm³ vs. 17.1 ± 29.8 cm³ (P = 0.035); PTV + 2 cm, 5.6 ± 14.5 cm³ vs. 26.8 ± 46.9 cm³ (P = 0.035); and PTV + 3 cm, 9.0 ± 17.4 cm³ vs. 36.5 ± 63.2 cm³ (P = 0.021). The overlapping volumes of the small intestine in the prone and supine positions are shown in Fig. 3. The small intestine was more than 1 cm away from the PTV in 12 patients (70.6%) in the prone position and 10 patients (58.8%) in the supine position.

The overlapping volumes of the rectum with PTV + 0.5 cm and PTV + 1 cm, and those of the bladder with PTV, PTV + 0.5 cm, PTV + 1 cm, and PTV + 2 cm were significantly smaller in the prone position. The overlapping volumes of the rectum and bladder are shown in Table 3. The overlapping volumes of the femurs with PTV + 2 cm (14.5 ± 9.1 cm³ vs. 11.9 ± 8.7 cm³, P = 0.0093), PTV + 3 cm (42.4 ± 14.8 cm³ vs. 37.5 ± 15.3 cm³, P = 0.0021), and PTV + 4 cm (77.0 ± 16.3 cm³ vs. 71.2 ± 17.8 cm³, P = 0.0039) were significantly larger in the prone position (Table 3).

In dosimetric comparison in the 11 appraisable patients, V40 of the rectum (47.9 ± 7.3% vs. 50.7 ± 5.8%, P = 0.046), mean dose of the bladder (39.5 ± 9.7 Gy vs. 42.0 ± 10.2 Gy, P = 0.0017), and V65 of the bladder (30.1 ± 11.7% vs. 33.5 ± 14.1%, P = 0.041) were significantly lower in the prone position (Fig. 4).

DVH parameters of the small intestine were as follows (prone position [mean ± SD] vs. supine position, respectively): maximum dose, 11.8 ± 11.8 Gy vs. 9.0 ± 9.3 Gy (P = 0.068); V15, 0.9 ± 2.6 cm³ vs. 0.4 ± 1.2 cm³ (P = 0.28). No part of the small intestine received over 45 Gy in the 11 patients.

DVH parameters of the rectum were as follows (prone position [mean ± SD] vs. supine position, respectively): mean dose, 42.9 ± 2.6 Gy vs. 43.5 ± 1.8 Gy (P = 0.21); maximum dose, 71.0 ± 0.6 Gy vs. 71.2 ± 0.5 Gy (P = 0.66); V70, 0.2 ± 0.2% vs. 0.3 ± 0.2% (P = 0.28); V65, 14.5 ± 2.6% vs. 14.8 ± 2.9% (P = 0.72); V50, 34.7 ± 6.0% vs. 36.9 ± 5.0% (P = 0.053).

DVH parameters of the bladder were as follows (prone position [mean ± SD] vs. supine position, respectively): maximum dose, 71.0 ± 0.7 Gy vs. 71.1 ± 0.4 Gy (P = 0.54); V70, 0.4 ± 0.7% vs. 0.3 ± 0.3% (P = 0.59); V50, 42.1 ± 14.2% vs. 45.6 ± 16.8% (P = 0.074); V40, 48.9 ± 15.2% vs. 52.3 ± 17.8% (P = 0.067).

DVH parameters of the sigmoid colon were as follows (prone position [mean ± SD] vs. supine position, respectively): maximum dose, 38.5 ± 28.1 Gy vs. 48.0 ± 24.9 Gy (P = 0.13); V65, 0.7 ± 1.3 cm³ vs. 1.3 ± 3.0 cm³ (P = 0.38); V50, 1.8 ± 3.2 cm³ vs. 2.9 ± 5.8 cm³ (P = 0.30); V40, 2.5 ± 4.3 cm³ vs. 4.0 ± 7.5 cm³ (P = 0.28).

DVH parameters of the femurs were as follows (prone position [mean ± SD] vs. supine position, respectively): maximum dose, 42.7 ± 5.1 Gy vs. 43.5 ± 2.0 Gy (P = 0.63); V40, 0.8 ± 0.9 cm³ vs. 1.0 ± 1.1 cm³ (P = 0.59); V30, 32.4 ± 17.5 cm³ vs. 41.7 ± 14.7 cm³ (P = 0.043). No part of the femurs received over 50 Gy in the 11 patients.

DVH parameters of the PTV were as follows (prone position [mean ± SD] vs. supine position, respectively): V107% (percent volume of the target volume receiving 107% of the prescribed dose), 0.1 ± 0.2% vs. 0.1 ± 0.1% (P = 0.77); HI, 1.23 ± 0.03 vs. 1.25 ± 0.04 (P = 0.21); CI, 1.43 ± 0.10 vs. 1.43 ± 0.11 (P = 0.97).

Bladder volume (r = 0.074; 95% CI, − 0.359 to 0.480; P = 0.74), CI (r = − 0.185; 95% CI, − 0.563 to 0.257; P = 0.41), the distance from the femur to the PTV (r = − 0.063; 95% CI, − 0.472 to 0.368; P = 0.78), time from surgery to the start of RT (r = − 0.234; 95% CI, − 0.597 to 0.208; P = 0.29), BMI (r = 0.409; 95% CI, − 0.015 to 0.709; P = 0.058), and age (r = 0.107; 95% CI, − 0.330 to 0.506; P = 0.64) did not have correlation with maximum dose of the small intestine (Fig. 5).

The small intestine is sometimes a dose-limiting organ in RT for pelvic malignancies. Yamashita et al. [13] reported seven patients who suffered from small bowel perforation after intracavitary radiation therapy in combination with external beam radiation therapy for uterine cervical cancer. In six of these patients the biological effective doses around the site of bowel perforation were 101.0–437.7 Gy (the dose distribution data was unavailable for the remaining patient) [13].

In patients with pelvic malignancies undergoing 3D-CRT or IMRT, the dose of the small intestine in the prone position has been shown to be lower than that in the supine position [15]. The same study also showed that the dose reduction was greater when the patient was in the prone position using a belly board. In a dosimetric study that investigated 10 rectal cancer patients who underwent IMRT [16], the prone position reduced the dose to the small bowel, while the dose distribution of the PTV was almost the same as that in the supine position. Another study of 3D-CRT for high-risk localized or locally advanced prostate cancer [17] showed that in the supine position, the volumes of the femoral heads and sigmoid colon that received a high dose of radiation were smaller than those in the prone position and no significant difference was found in the small intestine. In a recent study of the use of VMAT in the preoperative treatment of rectal cancer [18], the dose to the small intestine did not differ to a statistically significant extent between the prone and supine orientations; however, the setup reproducibility in the supine position was higher than that in the prone position. To the best of our knowledge, the present study is the first to show the effect of treatment positioning on the orientation of the OARs and the dose distribution during RT for prostate cancer in the postoperative setting.

In six patients (35.3%), the PTV included a fraction of the small intestine when they were in the supine position. They needed to be treated in the prone position in order to keep the small intestine away from the PTV. Even though the small intestine was more than 1 cm away from the PTV in the supine position in 10 patients (58.8%), the prone position seemed to be better because there was more space between the PTV and the small intestine, which enabled us to give priority to reducing the radiation dose that the other OARs would receive as well as the homogeneity of the dose distribution in the PTV. Actually, even though DVHs of the small intestine were not significantly different between in the prone position and in the supine position, the doses of the rectum and the bladder were lower in the prone position.

The femurs were nearer to the PTV when the patient was in the prone position, but DVHs of the femurs in the prone position was not inferior to those in the supine position. Although we administered oral gastrografin before CT scans, the enhanced regions of the small intestine did not affect the dose distribution.

The present study was associated with some limitations. In particular, we extended the CTV in the prone position by the same extent as the supine position, but the setup error in the prone position may bigger than that in the supine position. Even though we did not collect the data about the reproducibility of RT, we used a belly board and the patients underwent CT scans before every RT to compare them to the planning CT and correct the patient position if needed, which would contribute to improving the reproducibility in the prone position.

Our findings suggest that the volumes of the small intestine, rectum, and bladder in or near the PTV in adjuvant or salvage radiotherapy for prostate cancer after prostatectomy decreased when the patient was placed in the prone position on a belly board. DVHs of the rectum and the bladder were better in the prone position than in the supine position.