Introduction
Prostate cancer is the most common malignancy among male patients presenting to Oncology Department at this hospital which is in line with international statistics. Reference Stavrinides, Parker and Moore1 Being an underdeveloped nation combined with a lack of primary care, most patients with prostate cancer present late in our set-up. Hence among non-metastatic disease, high-risk prostate cancer is the most common presentation as opposed to western population where low-risk disease is more common. Reference Schroder, Hugosson and Roobol2 Risk grouping in non-metastatic prostate cancer was initially coined by D’Amico et al. which has been adopted ever since globally and with some modifications in most recent National Comprehensive Cancer Network prostate cancer management guidelines. Reference D’Amico, Whittington and Malkowicz3,4 Higher the risk group, more are the chances of distant metastases as well as pelvic lymph node failure. Ever since the publication of RTOG 9413 results, elective treatment of pelvic lymph nodes with radiation therapy has been practised with some scepticism as there was no overall survival (OS) benefit seen with this approach despite having disease-free survival advantage. Reference Roach, DeSilvio and Lawton5 Initial and long-term results of GETUG-01 have showed no survival benefit in a cohort of patients who received a prophylactic dose of radiotherapy to pelvic lymph nodes, that is, whole-pelvis radiotherapy (WPRT). Reference Pommier, Chabaud and Lagrange6,Reference Pommier, Chabaud and Lagrange7 Because of lack of standardisation regarding treatment of radiotherapy, STAMPEDE trial protocol did not mandate WPRT or prostate-only radiotherapy (PORT) and left it to organisational or individual decision. 8
Primary lymph node drainage area for prostate includes proximal external iliac, obturator, internal iliac, peri-rectal and pre-sacral lymphatic regions. Reference Breyer, Greene, Dall’Era, Davies and Kane9 The aim of treating these lymph nodes electively is to eliminate the micrometastases harbouring in them. Reference Pommier, Chabaud and Lagrange6 WPRT was performed routinely at our hospital since long until results from RTOG 9413 and GETUG-01 were available. Since then, there is a trend to omit pelvic lymph nodes from the radiotherapy treatment volume to avoid unnecessary side effects and a gradual move towards PORT. With the availability of 3D techniques and the use of the latest radiotherapy techniques like intensity-modulated radiotherapy (IMRT), dose escalation for better disease control was the preferred approach. While IMRT helped in dose escalation, it also helped in decreasing doses to organ-at-risk (OAR) which resulted in lesser toxicity from WPRT. The dose escalation for prostate stems from the concept of lower α/β ratio for prostate cancer, which has been proposed at 1·5–3 Gy range. Reference Onjuka, Uzan, Baker, Howard, Nahum and Syndikus10–Reference Tree, Alexander, Van As, Dearnaley and Khoo12 With this low α/β ratio, hypofractionation is hypothesised to enhance the therapeutic ratio by increasing local control without increasing the side effects. With better techniques of targeting the prostate, interest in PORT with escalated doses increased significantly. At our institute, since the availability of 3D simulation techniques in 2013, an overwhelming majority of high-risk non-metastatic prostate cancer patients were treated with PORT with neoadjuvant and adjuvant hormone therapy for a total of 2–3 years.
The aim of this study is to analyse the data of patients treated with PORT at this institute and see patterns of recurrence, especially those who had recurrence in pelvic lymph nodes.
Materials and Methods
A retrospective analysis of prostate cancer patients treated with PORT from 2014 to 2017, at Radiation Oncology Department of this hospital was carried out. All patients with prostate cancer who received PORT were initially shortlisted. Patients having histopathological diagnosis of prostate adenocarcinoma with Gleason score (GS) 8–10, or prostate-specific antigen (PSA) more than 20 ng/mL or AJCC stage T2c or above were included in the study. Reference Amin13 Patients having any one of the above factors were classified as high-risk group as per NCCN guidelines version 2021·1. 4 A small cohort of patients with low burden metastatic disease (as per STAMPEDE trial criteria) who also received radical dose PORT were included in the study. Shortlisted patients had Hb > 10 g/dL, normal serum urea and creatinine, normal liver function tests and were Eastern Cooperative Oncology Group (ECOG) performance status 0–1. Any patients with prior history of another malignancy, prior radiotherapy or prior chemotherapy for another malignancy were excluded from the cohort. Patients having metastatic disease detected on CT scan or a bone scan combined with a PSA > 100 but were treated as metastatic prostate cancer without a biopsy, were excluded from the study.
All eligible patients underwent histopathological diagnosis either through Transrectal Ultrasound biopsy or were incidentally diagnosed following Trans Urethral Resection of Prostate for treatment of their lower urinary tract symptoms. Patients having GS of 8–10, who did not undergo prior staging investigations had their staging investigations done with a bone scan and CT scan of chest, abdomen and pelvis. MRI pelvis was also carried out if not done prior to biopsy. Pre-biopsy PSA was recorded and in patients where pre-biopsy PSA was not recorded, PSA was measured 4 weeks after biopsy. All patients were treated with gonadotrophin-releasing hormone (Gnrh) agonist leuprolide or goserelin. Bicalutamide 50 mg per day was advised at commencement of hormone therapy and was offered for at least 14 days prior to first dose of Gnrh agonists. Patients with metastatic disease went on to receive chemotherapy or abiraterone in first 12 weeks at the start of hormone therapy. Androgen deprivation therapy (ADT) was continued for 2–3 years for non-metastatic cohort and indefinitely for the patients with metastatic disease.
Radiotherapy was planned at an interval of 4–6 months after the start of hormone therapy to have maximal downstaging before local definitive treatment. Different radical dose schedules of radiotherapy were permitted. Contouring guidelines of CHIPP trial protocol were widely followed for radiotherapy planning purpose. Reference Dearnaley, Syndikus and Mossop14 Radiotherapy planning was done using IMRT or volumated arc therapy using ECLIPSE treatment planning software by VARIAN Inc. Daily cone beam CT scan (CBCT) was used for positional verification during treatment delivery.
At the end of radiotherapy, all patients were monitored with serial PSA reading done every 3–4 months for 5 years according to American Society of Therapeutic Radiation Oncology guidelines for localised prostate cancer. 15 After 5 years, patients were advised to have PSA repeated every 6 months for another 5 years. These readings were recorded and all those patients who had a PSA failure based on Phoenix definition of a rise of PSA of more than 2 ng/mL above the nadir value were further investigated with bone scan and CT scan only as choline or PSMA PET scans were not available back then. Reference Matthew, Abramowitz and Tiaynu Li16
Patients with rising PSA despite castrate levels of testosterone who were receiving ADT were reclassified as either metastatic castrate resistant prostate cancer (mCRPC) or non-metastatic castrate-resistant prostate cancer (nmCRPC) depending on whether metastases were detected or not, respectively. Patients who had completed ADT were restaged and classified as metastatic hormone-sensitive prostate cancer (mHSPC) if metastasis was detected. Primary endpoint was pelvic lymph node failure rate. OS and failure-free survival (FFS) was also calculated using the Kaplan–Meier survival analysis using SPSS version 20.
Results
Two-hundred thirty-six prostate cancer patients were treated during the specified time period at this institute. A total of 105 prostate cancer patients fulfilling the inclusion criteria were retrospectively analysed. Mean age was 68·8 years (SD: 6·18). Median follow-up time was 62 months (Range: 18–94). Approximately 9 out of 105 patients had metastatic disease while only one patient had positive lymph node at diagnosis. Basic features are detailed in Table 1.
There were 40 patients in the cohort who had a PSA failure as per Phoenix definition. Only 3 out of these 40 had a metastatic disease in their pelvic lymph nodes either alone or in combination with metastasis at other sites as well. First patient had a GS of 5 and a T4 disease due to invasion of pelvic side wall at diagnosis. This patient was found to have a normal bone scan but a 2·2 cm lymph node in right internal iliac region without any other suspicious site on the scan for metastasis after being reinvestigated secondary to PSA failure. He was treated with stereotactic radiotherapy to the lymph node and responded well to that since he was alive at follow-up of 86 months without any on going hormone therapy.
Second patient who relapsed in pelvic lymph nodes had a GS of 10 and involvement of bilateral seminal vesicles at diagnosis and a PSA of 66·8 ng/mL. He had a PSA relapse at 13 months. His restaging scans showed multiple vertebral metastases along with left external and internal iliac lymph nodes. He progressed quite rapidly to subsequent lines of treatment and died of prostate cancer at 36 months post-diagnosis.
Third patient had a GS of 9 and pelvic floor muscle abutment at diagnosis. He also had low volume metastatic disease as per CHAARTED definition. Reference Sweeney, Chen and Carducci17 Thus, he was offered PORT. He had a PSA relapse at 29 months. There was no progression in bones but a solitary lymph node of 1·5 cm in short axis diameter was seen adjacent to bladder on the right side. Stereotactic radiotherapy was not suitable due to high bladder dose during prior PORT. Subsequent therapy with abiraterone showed good PSA response and patient is continuing on that till last follow-up at 46 months.
Sixty-five patients had adequately controlled PSA while 40 had a PSA failure at some point during their follow-up. Out of these 40 who had PSA failure, 4 were diagnosed as nmCRPC and were treated with ADT and enzalutamide. Twenty-two patients were reclassified as mCRPC. Ten patients had mHSPC while no metastases were detected in five patients but they remained sensitive to hormone therapy. Only one patient had isolated progression in pelvic lymph node without having any metastatic disease and was treated with stereotactic radiotherapy as mentioned above. FFS was 64% at 5 years. Median time to PSA failure in the 40 patients was 35·5 months (Range: 8–94). Kaplan–Meier curve for FFS is shown in Figure 1.
Seventeen patients had died during follow-up; out of which, 11 were due to prostate cancer and 6 patients succumbed to other causes. Seventeen patients in total were lost to follow-up. Reasons are not available for this high number of patients being lost to follow-up. Five year OS was 77%. Kaplan–Meier survival curve is shown in Figure 2.
Discussion
Radiotherapy to the prostate is crucial part of treatment for high-risk non-metastatic prostate cancer since long. Recently STAMPEDE trial has shown benefit of PORT in low-risk metastatic prostate cancer as defined by CHAARTED trial protocol. Reference Sweeney, Chen and Carducci17,Reference Parker, James and Brawley18 However, whether to include pelvic lymph nodes in high-risk non-metastatic prostate cancer inside the treatment field or not has long been studied with variable answers. Occult pelvic lymph node metastases, despite negative radiological imaging studies are common in high-risk prostate cancer. We retrospectively analysed our practise and found very few pelvic lymph node failures as mentioned above. Whether these results are in line with prior randomised trials or not will be discussed here.
RTOG 9413 was among the first of these trials to explore the issue in question. Reference Roach, DeSilvio and Lawton5 With more than 600 patients in each RT arm with or without pelvic RT, the nodal failure rates were exceptionally low at 1·3 and 2·5%, respectively. Eight percentage of patient population in GETUG-01 had local, nodal or distant failure in whole study population without any significant differences in arms with or without radiotherapy. Reference Pommier, Chabaud and Lagrange7 Although not separately mentioned, it amounts to approximately 17 cases of PORT arm which had 222 patients. Pelvic lymph node failures were not separately reported by GETUG-01. In our study, 2·8% patients, that is, 3 out of 105 cases developed failure in their regional lymph nodes. This bodes well with prior studies mentioned above. A very recently published POP-RT trial showed 15 regional recurrences in 114 patients (13·1%) in PORT arm as compared to 1 out of 110 patients treated with WPRT (0·9%). Reference Murhy, Maitre and Kannan19 This is the only known prospective randomised trial to have shown a significant difference in this regard.
Not all studies published on this subject had reported pelvic lymph node failure rates separately but have mentioned the lack of definitive benefit of WPRT in terms of OS, event-free survival or progression-free survival (PFS). A retrospective comparative study of 596 patients by Vargas et al showed no benefit of WPRT in terms of OS, rates of clinical failure or cause-specific survival. Reference Vargas, Galalae and Demanes20 While RTOG 9413 did show a significant 4-year PFS advantage without translating into OS advantage even with longer follow-up; there was no PFS advantage seen in GETUG-01 trial in favour of WPRT. Reference Roach, DeSilvio and Lawton5,Reference Pommier, Chabaud and Lagrange7 POP-RT did show significant biochemical failure-free survival, distant metastases free survival and disease-specific survival in favour of WPRT. Reference Murhy, Maitre and Kannan19 While RTOG 9413 and GETUG-01 do not recommend WPRT, POP-RT did conclude in favour of WPRT based on significant differences in above-mentioned endpoints. Reference Roach, DeSilvio and Lawton5,Reference Pommier, Chabaud and Lagrange7,Reference Murhy, Maitre and Kannan19
No OS benefit was seen in POP-RT, RTOG 9413 or GETUG-01 trials. Reference Roach, DeSilvio and Lawton5,Reference Pommier, Chabaud and Lagrange7,Reference Murhy, Maitre and Kannan19 Five year OS from our analysis has been compared with PORT arms of these trials in Table 2 and it shows that OS in our study is somewhat less than these studies. This may be explained by the fact that the population in our study had 10% patients with metastatic disease. Also that median PSA in our study was markedly higher than all three randomised trials in discussion as shown in Table 2. Eight out of 17 patients were lost before reaching median follow-up time. Variations in OS are also expected due to different inclusion criteria among these trials, for example, intermediate-risk disease patients being included in GETUG-01. Reference Pommier, Chabaud and Lagrange7 Forty-two patients in our study received 55 Gy in 20 fractions which was the dose in arm H of STAMPEDE. Reference Parker, James and Brawley18 However, this has since been replaced with 60 Gy in 20 fractions after publication of CHIPP trial results. Reference Dearnaley, Syndikus and Mossop14 Fifty-five Gy in 20 fractions is no more practised at our centre as a radical dose to prostate.
Abbreviation: LN, lymph node.
Since the publication of STAMPEDE arm H, it is well accepted that radiotherapy to the primary disease site is the corner stone of treatment. Reference Parker, James and Brawley18 Based on prior results from RTOG 9413 and GETUG-01, WPRT was not tested in arm H of STAMPEDE while the patients included in this trial were at the highest risk of lymph node metastases in the pelvis. Reference Parker, James and Brawley18 It would be interesting to know how the patients in arm H faired in terms of relapse in the regional lymph nodes. The final answer to the question of treating pelvic lymph nodes or otherwise is awaited from RTOG 0924; results of which are expected in late 2021 since the accrual has long been closed. It is testing the same question in unfavourable and high-risk non-metastatic prostate cancer. Some might argue in favour of WPRT since we have some positive results from POP-RT, but the number of patients in the trial were quite less for a phase –III trial (n = 224).
In this study, we tried to analyse outcome in our patient population treated with PORT and compare it with available data. The obvious weakness of the study is that it is a retrospective analysis and that too single arm. We wanted to compare it with data from patients treated at our centre who had received WPRT prior to the current trend but data were insufficient to make a comparison. Based on available results, prophylactic pelvic lymph node radiation may not be necessary due to very low rates of disease failure in pelvic lymph nodes, in patients with high-risk non-metastatic prostate cancer and metastatic prostate cancer who do not harbour lymph node disease at presentation till further Phase 3 data is available. Till then, prostate-only radiation can be practised as a safe option after thorough discussion with the patient.
Acknowledgements
None.
Financial Support
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Conflicts of Interest
The authors declare none.