Preliminary exploration of clinical factors affecting acute toxicity and quality of life after carbon ion therapy for prostate cancer

A total of 102 patients with prostate cancer were consecutively admitted from July 2015 to January 2018. The inclusion criteria were men with pathologically confirmed localized prostate cancer who received CIRT as the radical treatment. Patients with visceral or bone metastases [10], pelvic nodes metastases [11], postoperative radiotherapy [4], and recurrence after either surgery or radiotherapy [7] were excluded. Finally, 64 eligible patients participated in this study. Risk category was defined using the National Comprehensive Cancer Network (NCCN) prostate cancer guidelines [12]. This research had been approved by the institutional review boards of the SPHIC, and all patients provided written informed consent before admission.

Before CT simulation and every irradiation, patients were required to empty the rectum and fill the bladder. Patients are instructed to drink 330 ml of water 30–60 min prior to simulation and daily treatment. The organ at risk, including the rectum and bladder, was contoured on the basis of the Male Radiation Therapy Oncology Group (RTOG) Normal Pelvis Atlas [11]. The clinical target volume (CTV) was defined as the prostate alone for low-risk prostate cancer, prostate plus proximal 1–2.5 cm of seminal vesicles for intermediate- to very high-risk localized prostate cancer. The planning target volume was calculated from the CTV using the lateral margins of 10 mm as well as an anterior and posterior margin of 5 mm. Pelvic node radiation therapy was not used. Patients enrolled before January 2016 were treated by 66 photon Gray equivalent (GyE) (physical carbon ion dose [Gy] × RBE, which was assigned to be 3.0 at the distal part of the spread out Bragg peak) in 24 fractions, and by 59.2–60.8 GyE in 16 fractions thereafter. The CIRT was delivered once a day, 5 days a week.

Hormonal therapy was not administrated to low-risk patients, while intermediate-risk patients recommended to receive at least 6 months of hormonal therapy. And high-risk and very high-risk patients recommended for at least 2 years hormonal therapy.

Before CIRT initiation, demographic characteristics, international prostate symptom score (IPSS) and past medical history including transurethral resection of prostate (TURP) and hemorrhoids was recorded. Given the IPSS symptom questions were responsive to lower urinary tract symptoms [13], we divided the patients into 2 groups: IPSS ≤7 (mild symptoms) and IPSS ≥8 (moderate/severe symptoms), as previously described [14, 15].

The urinary-related adverse events, such as obstruction and cystitis, were defined as genitourinary (GU) toxicity, whereas the bowel-related adverse events were defined as gastrointestinal (GI) toxicity. Acute toxicity was evaluated using the Common Toxicity Criteria Adverse Event 4.03. Late toxicity was assessed by the RTOG/European Organisation for Research and Treatment of Cancer (EORTC) Late Radiation Morbidity scoring scheme [16].

The Expanded Prostate Cancer Index-Composite (EPIC) is a well-established patient-reported outcome questionnaire to monitor QOL among prostate cancer survivors. The 26-item version of EPIC, also known as the EPIC-26, contains symptom-based domains, such as urinary incontinence, urinary irritation/obstruction, sexual and bowel. The EPIC-26 domains, scoring from 0 (poorest) to 100 (best), can be tracked over time to understand symptom burden, functional outcomes and the impact of adverse effects [10]. The EPIC-26 Chinese version [17] was used in our study.

The QOL assessment was performed at the following 6 times: before CIRT (baseline), immediately after treatment finished, and at 3 months, 6 months, 1 year and 2 years after CIRT. The assessment was conducted in various ways, including face-to-face interviews, letters and online questionnaires; however, every patient had at least one face-to-face interview QOL assessment by an interviewer with a psychology background. The castration conditions were asked at each follow-up.

SPSS version 23.0 software was used. Nonparametric test and chi-square test were performed to assess patient characteristics and QOL scores between two prescribed dose groups. First, generalized estimating equations [18] were used to evaluate how QOL changed over time compared with baseline. 1/2 standard deviation (SD) from the baseline score, defined as the minimally important change [19], was also used in the quality-of-life. Then, if the QOL score at the last follow-up was lower than the baseline score, we thought the quality-of-life was declined. Next, Wilcoxon’s test and Fisher’s exact test were used for the univariate analysis, and a multivariable logistic regression was performed to correlate clinical parameters with occurrence of acute GU toxicity and reduced QOL. Variables that were statistically significant on univariate analysis (P < .10) were then analyzed in a multivariate regression model [20, 21]. P values less than.05 were considered significant. Age was considered as a confounding factor [22].

Table 1 shows the characteristics of the total 64 patients. There was no significant difference in baseline characteristics between two groups. All the patients received hormonal therapy except for 3 low-risk patients. Fifty-seven (89.1%) patients had a senior high school education or above, implying that they could provide a clear and precise description of their perceptions. Forty-six (71.9%) men were treated with 59.2–60.8 GyE/16 fx, whereas 18 men (28.1%) received a prescription dose of 66 GyE/24 fx.

Until January 2018, the median follow-up was 19 months (range, 3–33 months). 31 (48.4%) patients were not castrated at last follow-up and the reasons were low risk, completed hormonal treatment and intolerable drug toxicity.

The questionnaire response rate at each follow-up was over 85% (Table 2). Figure 1 illustrates how QOL changed over time in the domains of urinary irritation or obstruction, urinary incontinence, bowel and sexual function for all patients. The urinary incontinence scores for all patients were unchanged from baseline through 2 years of follow-up at all time points, whereas urinary obstruction/irritation was significantly poorer than baseline at the end of treatment (mean score change ± SD, − 7.92 ± 1.76; p < .001) but then recovered or was even superior to baseline in subsequent follow-up. At the end of treatment, when urinary function was at its poorest, patients reported “a moderate or big problem” mostly in urinary frequency (41.4%) and a weak stream (20%); however, these issues subsequently greatly resolved (Table 3). As for bowel function, although two dose groups’ bowel QOL scores differed at 3- and 6-month follow-up, they showed no significant difference in subsequent visit (Fig. 2). For all patients’ bowel QOL, the QOL scores maintained stability in the long term, but at the 2-year follow-up it demonstrated clinically relevant decrease (− 5.83 ± 2.74; p = .057). Some 32 (50%) patients had poorer bowel function at last follow-up. Nevertheless, after 6 months of follow-up, the number of patients who reported moderate or large problems with bowel function was zero. When it came to sexual QOL, the scores remained persistently stable after CIRT, but the baseline score was low (21.7 ± 18.5), with over 70% of patients complaining of multiple poor sexual functions (Table 3).

The occurrences of acute Grade 1 and 2 GU toxicity were 13 (20.3%) and 7 (10.9%), respectively, and the most common GU toxicity was urinary frequency (Grade 1 and 2: 8 and 7, respectively). There were also 5 cases of Grade 1 hematuria and 3 cases of Grade 1 dysuria. None of the patients developed grade 3 or higher acute GU toxicity. There were no acute GI toxicity occurring. With regard to late toxicity, there were 2 (3.1%) cases of Grade 1 and 1 (1.6%) case of Grade 2 cystitis for GU toxicity, with no late GI toxicity observed.

Table 4 shows the clinical factors that might affect the QOL after CIRT. TURP was a risk factor that predicted a decline in urinary QOL. Patients with a history of TURP had a 5.7-fold higher probability of poorer irritation/obstruction QOL and a 14.7-fold higher probability of impaired urinary incontinence QOL. Acute GU toxicity, IPSS ≥8 and castration status had no significant association with the urinary QOL decline after CIRT.

The history of TURP did not make a difference to the acute GU toxicity during the CIRT, though it was a remarkable risk factor of declined urinary QOL in the relative long term. Hormonal therapy made no difference, either. However, IPSS ≥8 made a significant difference and it predicted a 5.3-fold higher probability of acute GU toxicity during CIRT (Table 4).

When focusing on bowel QOL, the baseline score was a significant influence factor, with a higher baseline score tending to indicate a decrease in subsequent QOL. Though the discrepancy was not significant in the univariate analysis, it was significant in the multivariate analysis (p = .035). Castration status was a potential risk factor for declined bowel QOL with an odds ratio (OR) of 2.8, but it did not reach significance after multifactorial adjustment (p = .079). Neither aspirin therapy nor hemorrhoid history were significantly related to impaired bowel QOL.

As for sexual QOL, castration status was a significant risk factor. Patients maintaining castration until the last follow-up had a 12.7-fold increased risk of impaired sexual QOL after adjustments for age and baseline score.

All domains showed no significant difference between the two prescribed doses.