Moderate hypofractionated helical tomotherapy for older patients with localized prostate cancer: long-term outcomes of a phase I–II trial

Prostate cancer (PC) was one of the most common malignant tumors in men. PC therapy should theoretically benefit from hypofractionated radiotherapy (HFRT) due to its low α/β value which may be even lower than surrounding late-response tissues and organs.1, 2, 3 Our previous report had shown that HFRT was efficient and well tolerated in 67 PC patients with low incidences of severe acute toxicity complications.4

HFRT has gradually become the trend of treatment in many solid tumors, and its advantages, compared to conventionally fractionated radiation therapy (CFRT), are mainly reflected in higher level of biological effective dose (BED) with a lower total dose/fewer fractions, and shorter treatment course.3 Thus, without a significant increase of radiation related toxicities, it can effectively save medical resources and bring potential economic benefits. Studies evaluating HFRT regimen in prostate cancer reported that it was not inferior to conventional CFRT in efficacy and was not associated with increased late toxicity.5, 6

In addition, according to Roach et al.,7 in patients with a risk of lymph node (LN) involvement, due to the need of preventive irradiation for pelvic lymph nodes, simultaneous modulated accelerated radiation therapy is recognized as the most appropriate RT method. At present, two HFRT regimens, including moderate hypofractionation and ultra-hypofractionation, are applied to localized PC. Some randomized controlled trials (RCTs) have confirmed the safety and efficacy of moderate HFRT compared to CFRT, in particular comparable rate of grade ≥ 2 adverse events6 or grade ≥ 3 late genitourinary and gastrointestinal toxicity.8 However, few results from RCTs are available to support the application of ultra-hypofractionation in high-risk tumors.

Based on the foregoing, the aim of our study was to compare the long-term outcomes and toxicities between the moderate HFRT regimens and to further investigate the feasibility of shorter-duration, moderate HFRT for high-risk prostate cancer. After > 7 years follow-up, in this paper we report treatment outcomes and late toxicities.

This nonrandomized, single center, prospective Phase I–II trial showed that older patients with localized prostate cancer treated with 76 Gy/34F or 71.6 Gy/28F both had similar PFS, BFFS, PCSS, and OS for more than 7 years follow-up. Two moderate hypofractionated therapy regimens shared mild high-grade late GU and GI toxicities, indicating well toleration. For the risk of pelvic lymph nodes involvement (LNI) > 15%, 50.4 Gy/25F prophylactic pelvic node irradiation showed no difference in PFS, BFFS, PCSS and OS, compared to similar risk patients who did not undergo further treatment.

It is well known that local control of malignant tumors can be improved by increasing the BED, which is usually related to three factors: total dose, fractionated dose and total treatment time. In the 1990s, under the background of two-dimensional (2D) radiation therapy and three-dimensional conformal radiotherapy (3DCRT), increasing the total dose became a hotspot of research, and there were also studies on dose-escalated radiation therapy in PC patients. Many studies had confirmed that increasing the prescription dose to 78–80 Gy in conventional fraction could significantly improve BFFS and reduce PC-related mortality.14, 15, 16 According to the National Comprehensive Cancer Network (NCCN) guidelines of 2005, the prescription dose for low-risk patients should reach to 70– 75 Gy, while for patients with intermediate- and high-risk that should be up to 75–80 Gy in conventional fraction.17 However, as the dose was further increased, the incidence of severe GU and GI toxicities would increase significantly, which limited the benefit from dose-escalation with CFRT. In the Dutch multicenter phase III study5, the cumulative incidence of ≥ grade 2 GI toxicities was 35% in the 78 Gy group and 25% in the 68 Gy group (p = 0.04). MRC RT01 multicenter phase III study18 in UK confirmed that the incidence of late bowel toxicity in the 74 Gy group was higher than that in the 64 Gy group (33% vs. 24%) according to the RTOG (grade ≥ 2) scale within 5 years from starting treatment.

Intensity modulated radiation therapy (IMRT), developed on the basis of 3DCRT, can deliver a high dose to the target volume and effectively protect the surrounding OARs. RTOG-0126 study19, 20, which enrolled 1532 PC patients with intermediate-risk, confirmed that compared with 3DCRT, IMRT could significantly reduce ≥ grade 2 acute GU and GI toxicities. Univariate and multivariate analyses also showed a lower incidence of ≥ grade 2 late GI toxicities in IMRT group.21 Cahlon et al.22 increased the prescription dose to 86.4 Gy/48F (BED3 = 138.24 Gy) using IMRT technique and achieved excellent results. The incidence of ≥ grade 2 GI and GU toxicities was 3.8% and 15.1%, and the 5-year BFFS was 98%, 85% and 70% in the low-, intermediate-, and high-risk groups, respectively. In a meta-analysis published in 2009, Viani et al.23, 24 pointed out that prescription dose was directly proportional to biochemical control for patients with localized PC and high-dose radiation therapy was superior to conventional-dose radiation therapy. However, it was not easy to achieve high dose with 3DCRT technique, and IMRT showed its advantages.

Figure 4

At the beginning of this century, many studies have shown lower α/β value of PC than many common cancers, even lower than late-responding tissues. Therefore, when the total dose remained the same, increasing fractionated dose could effectively kill PC cells, with limited increased toxicities. Based on this theory, studies of HFRT on prostate cancer became the hotspot from then till now.25 Arcangeli et al.26 first reported the long-term results of a phase III study. The 70-month BFFS in patients with high-risk PC in the HFRT group (62 Gy/20 F) and CFRT group (80 Gy/40 F) was 85% and 79% (p = 0.065), respectively. The 10-year BFFS was 72% in HFRT group and 65% in CFRT group (p = 0.148), and 95% and 88% for the 10-year PCSS (p = 0.066), respectively.27 Although no significant difference was detected between the two groups, the study revealed that hypofractionation was a significant prognostic factor for BFFS and PCSS. Kupelian et al.28, 29, 30 used IMRT technique with a prescription dose of 70 Gy at 2.5 Gy per fraction, and the initial results showed low incidence of adverse reaction without ≥ grade 2 late toxicities after 18 months follow-up. The 5-year BFFS was 82% (95% CI: 79%–85%) and the incidence of ≥ 2 grade late GI and GU toxicities was only 6% and 7%, respectively. In 2013, Pollack et al.31 first reported the results of a phase III trial using hyporfractionated IMRT technique, in which two fractionation regimens of 76 Gy/38 F and 70.2 Gy/26 F were compared. The 5-year BCDF was 21.4% and 23.3% (p = 0.745) without significant difference, and the incidence of late toxicities was similar in the two groups. The results were confirmed in the subsequent phase III trials (CHHiP, HYPRO, PROFIT and RTOG-0415).6,8,21,32 This study was a dose-escalating trial using two hypofractionation regimens (BED3 was more than 130 Gy, and BED1.5 was about 190 Gy), and the results were similar to the previously published literature in terms of late toxicities or survival, with the 7-year PFS, BFFS, PCSS, and OS for all the patients being 71.6%, 77.6%, 91.9%, and 77.0%, respectively, without significant differences between the two regimens. The meta-analysis of Yin et al.33, in which seven of 365 studies fulfilled inclusion criteria with 8156 participants, provided reliable evidence that moderate HFRT decreased BF rate, while did not improve OS. Compared with CFRT, HFRT with an increase in BED1.5 improved BFFS, and accordingly an increase in BED5 would result in elevated late GI and GU toxicities. Although those results did not show OS benefit from HFRT in PC patients, it is necessary to carry out more randomized trials with more samples and longer follow-up to achieve better prognosis for PC patients.

Another original intention of using HFRT for PC patients was to reduce GU and GI toxicities. In this study, we selected a scan thickness of 4 mm for IGRT, and only 0.015 Gy dose received by patients for each scanning, so even with daily image guidance, the cumulative dose was far less than actual prescription dose. After 7 years of follow-up, no radiation-induced second primary tumor was detected. Although different fractionation regimens, RT techniques, and evaluation systems had been used in different clinical trials, the results were not satisfactory. This may be associated with the special anatomical location of the prostate, which is close to bladder and rectum, and affected by the filling degree of the two organs. Because of the uncertainty of target location, high dose area would cover a certain volume of bladder and rectum. According to the traditional positioning methods, the CTV to PTV margin often required > 1cm, which would increase the incidence of adverse reactions. Relevant study suggested that reducing the CTV-PTV margin could reduce the normal tissue complications (NTCP) of rectum by up to 10%.34 Maund et al.35 showed that the reduction in NTCP for > 2 grade rectal toxicity of 0.7% corresponded with a 2 mm margin reduction for IMRT. Utsunomiya et al.36 used both IMRT and CTV-PTV margin reduction technique, and reported rectal NTCP < 5% when the dose was escalated from 70 to 78 Gy.

Three-dimensional image guidance technique is another revolutionary progress in radiation therapy (image-guided radiation therapy, IGRT). By monitoring and correcting deformation and displacement of the target and OARs, it provides effective help to improving irradiation accuracy and reducing the CTV-PTV margin. In a sub-study of CHHiP trial, 293 patients were secondly randomized and assigned to no-IGRT, IGRT-with standard CTV-PTV margins, or IGRT-with reduced CTV-PTV margins. Rectal and bladder dose-volume and surface percentages were significantly decreased by the reduction of CTV-PTV margins, and overall side effect profiles were acceptable in all groups but lowest with IGRT and reduced margins.37 In 2019, ESTRO ACROP (Advisory Committee on Radiation Oncology Practice) released the consensus guideline on the use of IGRT for localized PC, and daily on-line correction was preferred for CFRT and recommended in case of HFRT, with a reduction of the CTV-PTV margin to 4–6 mm. In this study, the TomoTherapy system had the image-guidance function with its MV-CT. Compared with KV-CT, the resolution of soft tissue was relatively poor, but the spatial resolution and the homogeneity of images were the same.38 With daily on-line image-guidance in our previous study, with a 5-mm margin in left-right and cranial-caudal directions, and a 3-mm margin in anterior-posterior direction, both the incidences of acute GU and GI toxicities were only 4.7%⁴, which was similar to the studies conducted by Murthy et al.39 and Schiller et al.40 using TomoTherapy system. Relevant studies and our data also showed that the imaging dose was generally between 0.01 Gy and 0.03 Gy, which significantly correlated with pitch and layer thickness.

At present, high-dose CFRT or moderate HFRT has been recognized as standard treatment for localized PC, but whether to perform pelvic lymph node irradiation for intermediate- or higher risk patients is still controversial. Since the Roach formula was used to clinically predict the probability of pelvic lymph node metastasis, most studies used LNI risk > 15% or > 30% as the reference value for preventive pelvic irradiation. Earlier retrospective studies all showed that intermediate- and high-risk patients could benefit from this irradiation. However, the level of evidence for these results was low due to differences in prescription dose, absence of ADT, and impact of confounding factors. Until now, there had been two phase III studies comparing results of either applying elective lymph node irradiation or not in patients with intermediate- and high-risk PC, and long-term follow-up results had been obtained. The GETUG-01 study41 showed that elective lymph node irradiation did not improve event-free survival or OS. The longterm update of RTOG 9413 study demonstrated neoadjuvant ADT plus whole pelvic radiotherapy improved 10-year PFS compared with neoadjuvant ADT plus prostate only radiotherapy and whole pelvic radiotherapy plus adjuvant ADT, albeit increased risk of grade 3 or worse intestinal toxicity.42 However, neither trial used IMRT technique or delivered doses that would be considered inadequate by today’s standards. The ongoing RTOG 09-24 trial using IMRT with increased dose might provide more conclusive evidence for the effects of pelvic node irradiation. In this study, the strategy of pelvic lymph node irradiation was different between the two groups. Twenty-five patients with LNI risk > 15% by Roach formula in Group-2 all received elective lymph node irradiation, while 23 patients with the same risk in Group-1 did not. The 7-year BFFS and OS were 62.1% and 68.2% in Group-1, and 68.8% and 82.6% in Group-2, without significant differences between the two groups, even for the 7-year BCDF which was 46.9% and 34.4%, respectively. At the same time, the incidence of late GU and GI toxicities for patients with LNI risk > 15% by Roach formula, either received elective lymph node irradiation or not also had no statistical difference. The reasons may be as follows: 1) High dose IMRT, combined with daily image guidance, targeting prostate and seminal vesicles would be sufficient in localized PC patients who have no clinical node involvement; 2) For older patients with PC, non-tumor factors had a greater impact on survival, which partly covered the impact of tumor itself; 3) This non-randomized study with small sample size might have resulted in biased statistical results.

In 2016, the American Joint Committee on Cancer (AJCC) established criteria to evaluate prediction models for cancer staging, with following works indicating high-risk for prostate cancer defined by a patient’s Gleason score, prostate-specific antigen level, and clinical AJCC T stage.10,43 In addition to that, intraductal carcinoma of the prostate44,45 and 22-Gene Genomic Classifier46 were reported as a prognostic factors for PFS, CSS, and OS in patients with high-risk prostate cancer. In line with those findings, clinical T stage > 3b in our study was an independent prognostic factor of PFS (HR 0.197, 95% CI 0.065–0.594, p = 0.004). As far as we know, patients with different age distributions may correspond to different survival outcomes in many cancers. Patients at extreme ages may not be able to accurately show the effects of intervention factors, so ≥ 70 or 75 years-old participants were usually underrepresented in most of the clinical trials. Unlike other cancers, prostate cancer often develops in patients with old age, and over 80% of the cases are diagnosed after the age 65. Syrigos et al.47 thoroughly reviewed the detailed evidence of prostate cancer in the older patients and concluded that age alone should not constitute an obstruction for optimal treatment administration. In a retrospective outcome analysis of radical radiotherapy for PC patients, comparing patients above or below the age of 80, no difference in 5-year BFFS, distant metastasis-free survival (DMFS), or PCSS was detected.48 A meta-analysis showed that radiation therapy seemed to be associated with a reduction in the risk of death in patients aged 80 or above with clinically localized PC compared to observation.49 These studies all showed that the treatment for older patients with prostate cancer should be more active. Besides that, excellent outcomes were also reported in these patients with HFRT.50, 51, 52 However, there have been few randomized studies focusing on the outcome of HFRT in older patients with prostate cancer. In this study, the median age was 80 years and 78 years in the two groups, respectively, and both showed satisfied outcomes and low incidence of toxicities.

This study has several limitations. Firstly, the outcomes are from a single institution, and the sample size of the trial is relatively small, resulting in a large proportion of failures probably caused by non-tumor factors of high-risk patients. Secondly, robust multivariate analysis was not feasible due to a small number of observed toxicity events, and multiple analysis of clinical factors associated with toxicity needed further investigation. Clinical trial data from other centers are needed to evaluate the identified association of clinical parameters with toxicity.

In conclusion, two moderately hypofractionation regimens, 76 Gy/34 F and 71.6 Gy/28 F, delivered with daily image-guided HT technique, were well tolerated in older aged prostate cancer patients with minor severe late toxicities and satisfied survival, and prophylactic pelvic lymph node irradiation might not be necessary. A phase III trial is needed to explore the optimal hypofractionation regimen and the necessity of prophylactic pelvic node irradiation in older aged patients with localized prostate cancer.