Long-term outcome of a moderately hypofractionated, intensity-modulated radiotherapy approach using an endorectal balloon for patients with localized prostate cancer

Dose-escalated radiotherapy is an established treatment for localized prostate cancer. Technical advances in radiation delivery, including three-dimensional conformal radiotherapy (3D-CRT) and, more recently, intensity-modulated radiotherapy (IMRT) have simultaneously enabled dose escalation and enhanced bladder and rectal sparing. These developments have led to demonstrable gains in therapeutic ratio through improved disease control rates and concomitant reductions in acute and chronic adverse events [1, 2].

Evidence from multiple retrospective and prospective series of patients with localized prostate cancer confirmed the theoretical benefits of dose escalation. A mature phase III trial from M.D. Anderson Cancer Center revealed decreases in biochemical and clinical disease progression (including distant metastases) rates for patients treated with an isocenter dose of 78 Gy compared with 70 Gy [3]. Similarly, a randomized trial by Zietman et al. [4] demonstrated benefits from proton boost escalation of 79.2 vs. 70.2 Gy, and a Dutch randomized trial revealed improved freedom from failure for patients receiving 78 vs. 68 Gy [5]. Prospective dose escalation data from Memorial Sloan Kettering Cancer Center suggested similar clinical benefits, including reduced distant failure independent of short-term androgen deprivation therapy (ADT) [6].

Collectively, the above studies reflect progressive gains using conventional fraction sizes of 1.8–2.0 Gy and are validated by updated National Comprehensive Cancer Network (NCCN) guidelines [7]. However, the optimal radiation fractionation regimen for localized prostate cancer is unclear as a result of mounting laboratory and clinical evidence suggesting that the alpha/beta (α/β) ratio for most prostate cancers lies between 1 and 3 Gy [8–10]. This hypothesis provides a scientific rationale for hypofractionation [8–11] and, if verified, would have significant therapeutic implications including improved patient convenience and cost savings by virtue of reduced number of fractions. Data from a handful of prospective trials [12–15] are consistent with the hypothesis. However, significant heterogeneity exists among reported hypofractionation regimens [12–15].

Herein, we report our institutional outcomes using moderately hypofractionated IMRT, and a rectal balloon, with emphasis on long-term biochemical control and treatment-related adverse events in patients with localized prostate cancer.

We have presented the long-term outcomes of a large cohort of patients with IMRT-treated prostate cancer. Study strengths include mature follow-up and homogeneous treatment techniques, as all patients underwent full-course IMRT and were treated in the prone position with a rectal balloon using a uniform, moderate hypofractionation regimen, and using consistent treatment planning and delivery methods at a single facility. The results demonstrate good disease control and acceptable rates of adverse events.

We opted to use an endorectal balloon for our patients due to several advantages. The balloon provides a constant rectal filling that immobilizes the prostate by pushing it towards the pubic bone. In addition, the balloon pushes the dorsal wall of the rectum away from the prostate, resulting in improved rectal sparing and reduced adverse events. Our group has previously published on the reduced dose to the rectum at the air-tissue interface between the balloon and the rectal wall [19]. However, the rectal balloon pushes the anterior wall of the rectum closer to the prostate; therefore, there is concern that because part of the rectum is close to the high dose region, there might actually be an increase in rectal adverse events.

Early hypofractionation randomized trials used 3D-CRT technique with relatively low total doses ranging from 52.5 to 55 Gy in 20 fractions (NTD_2Gy, 61.5–66.8 Gy₂) [28, 29]. As expected, treatment outcomes were suboptimal, with 5-years bRFS ranging from 40% to 55.9%. Other 3D-CRT series, however, employed more aggressive fractionation regimens, with NTD_2Gy values ranging from 77.1 to 83.7 Gy₂ and have reported improved outcomes (5-year bRFS rates generally above 80%, with some exceptions) despite including patients with high-risk disease [14, 30–32].

A phase I/II trial by Kupelian et al. [12] was a large prospective study of hypofractionated IMRT and prescribed 70 Gy in 28 fractions (NTD_2Gy = 80 Gy₂). At a median follow-up of 45 months, the authors reported 5-year bRFS rate of 83% for the entire cohort, and 94%, 83%, and 72% for low-, intermediate-, and high-risk cohorts, respectively. Severe (Grade ≥ 3) late rectal and urinary adverse events were uncommon. Results from two additional randomized trials have been recently reported. Pollack et al. [13] tested a hypofractionation regimen delivering 70.2 Gy in 26 fractions (NTD_2Gy = 84.4 Gy₂). Among 151 patients and with a median follow-up of 68.4 months, the observed 5-year bRFS rate was 76.7%. The 5-year rates of Grade ≥ 2 GI and GU adverse events were 18.1% and 21.5%. Similarly, Kuban et al. [15] employed 72 Gy in 30 fractions (NTD_2Gy = 80.2 Gy₂) and reported, at a median follow-up of 4.6 years, a 5-year bRFS rate of 96% for the 102 patients in the hypofractionation group. There were 9 patients with Grade 2 and 2 with Grade 3 GI adverse events, for 5-year actuarial rates of 11% and 3%. Overall, the 5-year rate of Grade ≥ 2 GU adverse events 19%.

More recently, three randomized, phase III trials published mature results comparing conventionally fractionated radiotherapy with hypofractionated radiotherapy. The largest was CHHiP [32], which was a non-inferiority study that randomized 3216 patients with localized prostate cancer to conventional (74 Gy delivered in 37 fractions) or one of two hypofractionation regimens (60 Gy in 20 fractions or 57 Gy in 19 fractions, NTD_2Gy = 77.1 or 73.3 Gy₂, respectively) delivered via intensity-modulated techniques. The observed 5-year bRFS rate was 88.3% in the 74 Gy group vs. 90.6% and 85.9% in the hypofractionation groups. The study found that 60 Gy was non-inferior to 74 Gy in terms of bRFS, but non-inferiority could not be claimed for 57 Gy compared with 74 Gy. The rates of Grade ≥ 2 GI and GU adverse events were 13.7% and 9.1% in the 74 Gy group, 11.9% and 11.7% in the 60 Gy group, and 11.3% and 6.6% in the 57 Gy group, respectively. The authors recommended hypofractionated radiotherapy as a new standard of care for localized prostate cancer.

Another large randomized, phase III trial [33, 34] was conducted at 7 Dutch centers. It enrolled 804 patients with localized prostate cancer and randomized them to either hypofractionated radiotherapy of 64.6 Gy (19 fractions of 3.4 Gy, 3 fractions per week, NTD_2Gy = 90.4 Gy₂) or conventionally fractionated radiotherapy of 78 Gy (39 fractions of 2 Gy, 5 fractions per week). The 5-year bRFS rate was 80.5% for patients in the hypofractionation group and 77.1% for those in the conventional fractionation group. The 3-year rate of Grade ≥ 2 GU adverse events was 39.0% in the conventional fractionation group and 41.3% in the hypofractionation group. In addition, there was a significant increase in the 3-year rate of Grade ≥ 3 GU adverse events in the hypofractionation group compared with that in the conventional fractionation group (19% vs. 12.9%). The 3-year rate of Grade ≥ 2 GI adverse events was 17.7% in the conventional fractionation group and 21.9% in the hypofractionation group. The authors concluded that their hypofractionation regimen could not be regarded as the new standard of care.

The third trial with recent results was conducted by NRG [National Surgical Adjuvant Breast and Bowel Project (NSABP), the Radiation Therapy Oncology Group (RTOG), and the Gynecologic Oncology Group (GOG)] Oncology (0415) [35], which randomized 1092 men with low-risk prostate cancer to hypofractionated radiotherapy (70 Gy in 28 fractions, NTD_2Gy = 80 Gy₂) versus conventionally fractionated radiotherapy (73.8 Gy in 41 fractions, NTD_2Gy = 69.6 Gy₂). 3D-CRT or IMRT were allowed. The estimated 5-year bRFS rate was 85.3% in the conventional fractionation group and 86.3% in the hypofractionation group. There was an increase in late Grade ≥ 2 GI and GU adverse events in the hypofractionation group (22.4% and 29.7%, respectively) compared with the conventional fractionation group (11.4% and 20.5%). The difference in the rate of late Grade 2 GI adverse events reached statistical significance. The authors concluded that hypofractionation was non-inferior to conventional fractionation.

These recently published results give credence to the idea that hypofractionated radiotherapy is an effective treatment for prostate cancer in terms of disease control. There is concern, however, over the possibility of increased late GI and GU adverse events as compared with conventional fractionation, although overall rates remain low. This concern, along with the heterogeneity in the various hypofractionation regimens, has led some clinicians to give pause before offering hypofractionated radiotherapy as standard of care therapy. However, the recently published data likely does obligate practitioners to discuss the risks and benefits of hypofractionated radiotherapy with their patients, particularly those who are interested in a shortened course of radiotherapy.

At its inception, our program sought to safely improve biochemical control rates by delivering an increased biological effective dose to the prostate target volume while maintaining a constant number of fractions relative to the conventional doses of 66–70 Gy which predated the modern dose escalation era. The long-term treatment outcomes using a mean dose prescription of 76.7 Gy in 35 fractions (NTD_2Gy = 80.9 Gy₂) demonstrate the safety and efficacy of this fractionation regimen by virtue of excellent long-term bRFS and low rates of severe late GI and GU adverse events, which falls in line with the recently published randomized data.

We recognize that our study has important limitations, including its single-institution, single-arm, retrospective design, lack of detailed dose-volume histogram analysis, omission of image-guided delivery techniques, and potential bias in the use of hormonal therapy, particularly among low-risk patients. Nevertheless, the more recent published studies reported similar outcomes and give some support to our data.

Long-term bRFS and adverse event outcomes in patients with localized prostate cancer treated using a moderately hypofractionated IMRT regimen are encouraging. Based on this large single-institution study, in combination with the recently published data from three randomized clinical trials, moderate hypofractionation represents a safe, efficacious regimen in the treatment of localized prostate cancer that should be discussed with patients interested in a shortened course of treatment.