Comparing robotic-assisted to open prostate surgery: 2-year outcomes

[Wendy_Winnall]

Two years ago, a landmark Australian study of prostate surgery was published in a top journal. Conducted in Queensland, this randomised controlled trial compared the two main types of prostate surgery. Twelve weeks after surgery, there were no obvious differences between the sexual side effects, urinary side effects and success in treating in treating the cancer. At the time, many considered that 12 weeks was not enough time to see any difference. The 2-year results now are in!

There are two main types of surgery for removing a prostate gland. The robotic-assisted laparoscopic prostatectomy is a "keyhole" surgery, where instruments are inserted through smaller incisions. The surgery is not actually conducted by robots. Rather, the surgeon controls the surgical instruments remotely, via the robotic system. This allows them very fine control over how the instruments are used. This type of surgery is herein referred to as robotic-assisted surgery. The second surgery type is open radical retropubic prostatectomy. This older style of surgery involves a large incision in the abdomen though which the surgery is performed. It's referred to here as open surgery. 

Robotic-assisted surgery for prostate cancer was first performed in 2001 and introduced into Queensland in 2009. In Australia, the robotic-assisted surgery quickly became very popular. Now the majority of prostate surgeries are done by the robotic-assisted technique in many developed nations.

Upon its introduction to Australia, robotic-assisted prostate surgery was known to be safe and effective. There were many claims made by various parties that it was superior to the open surgery in terms of success (preventing return of the cancer), side effects and recovery. However, unlike the introduction of a new medicine, no large randomised controlled trials were necessary to make these claims. Although a number of good studies have compared the surgery types and found in favour of the robotic-assisted surgery, these were not randomised controlled trials - the gold standard of evidence. The randomised controlled trial design is the only one that can give an unbiased comparison of the two surgery types.

There has only been one phase 3 randomised controlled trial comparing these two surgery types. This landmark trial was conducted in Australia. Australian scientists seized the opportunity to test this issue properly in Queensland, when the first robotic system was delivered to the state. The first robotic system was installed in 2009 at the Royal Brisbane & Women's Hospital. At this time, the surgery was relatively new. Australian men with prostate cancer were happy to allow the trial co-ordinators to randomly choose their surgery type. This randomisation was essential, and made the trial possible. Now that robotic-assisted surgery has become so popular, it's unlikely that another randomised controlled trial could be conducted in a developed nation. Men would want to choose their surgery type, rather than have it assigned by the trial.

The Australian surgery trial was run by Prof Frank Gardiner, eminent clinician and researcher from the Royal Brisbane & Women's Hospital and University of Queensland. His trial was designed to ask if there were any detectable differences in the success and side effects of the two main types of prostate surgery. Robotic-assisted and open surgery were compared in Queensland patients with localised prostate cancer. Men were recruited to the trial if they had localised prostate cancer and had chosen surgery as a treatment. 157 men had robotic-assisted surgery and 151 had open surgery to remove their prostate glands. Their recovery, physical and mental health after surgery were closely monitored over 2 years (so far).

12 weeks after surgery (results from 2016)

The 12 week results were published 2 years ago and were the subject of a research blog at that time. Here is a summary of the results from 12 weeks after surgery:

1. 10% of men who had open surgery and 15% who had robotic-assisted surgery had signs of cancer cells being left behind, but this difference was not statistically significant.
2. No significant difference was found in urinary function between the two types of surgery (by 12 weeks).
3. No significant difference found between the two types of surgery for sexual function (by 12 weeks).

There were no differences found for bowel functioning, cancer-specific or psychological distress between the surgery types, all measured by specific surveys.

There were differences in the level of pain experienced immediately after surgery. Patients who had the robotic-assisted surgery reported less pain during activity at 24 hours after surgery and 1 week later. Six weeks after surgery there were no differences in pain levels. A slightly lower score for physical functioning in the open surgery group was reported 6 weeks after surgery. This was no longer apparent by 12 weeks. The open surgery resulted in more blood loss, but no blood transfusions were necessary. Patients who had open surgery stayed in hospital for longer, but there was no significant difference in their time away from work. These differences are probably due to the keyhole style of surgery. Keyhole surgery is usually less painful to recover from than open surgery and has a lower chance of some complications, such as blood loss.

Publication of these results caused a lot of debate among clinicians (described below).

2-year update on progress

The 2-year update has been recently published in the journal Lancet Oncology. In this publication is data comparing the two groups of men at 6 months, 12 months and 2 years after their surgery:

1. PSA levels (indicating prostate cancer recurrence): 13 of the men who had open surgery and 4 of the men having robotic-assisted surgery suffered a rise in PSA levels (PSA>0.2 ng/ml). This was 9% of the men having open surgery and 3% of those having robotic-assisted surgery. A statistical test indicated that these differences were statistically significant. However, there was no difference found in the proportion of men for whom scans showed that their tumours had spread.

The higher proportion of men with rising PSA after open surgery was surprising, given that more men in the robotic-assisted surgery group had positive surgical margins (evidence that there was some tumour left behind after surgery). One possible reason was that more men in the robotic-assisted surgery group had radiotherapy after their surgery.

The study authors recommended caution in interpreting these results because the treatment of the two groups after the operation was different. What they really mean is - they are unsure if this statistical difference is real.

2. Urinary function: Surveys conducted at the 6-month, 12-month and 2-year marks found no significant differences in urinary function between the two groups of men. By 2 years, 91% of the men who had robotic-assisted surgery and 95% of the men who had open surgery had good urinary continence levels (did not require pads for incontinence).

3. Sexual function: Surveys conducted at the 6-month, 12-month and 2-year marks found no significant differences in sexual function between the two groups of men. 36% of the men who had open surgery had erections firm enough for intercourse more than half the time. 38% of the men having robotic-assisted surgery also achieved this level of erections. However, there was a significant difference detected between the groups 12 months after surgery. The men who had open surgery had greater use of tablets for sexual problems. By 2 years, 46% of men in the open surgery group had used a sexual aid in the past 4 weeks, compared to 37% in the robotic-assisted group.

There were no significant differences found between the groups for measures of quality-of-life (mental and physical), bowel function, cancer-specific distress, vitality and urinary symptoms within the 2-year follow-up.

This study also measured psychological distress using a survey. There was no difference in the proportion of men experiencing psychological distress between the two groups. But this analysis did show that about 1 in 5 men showed signs of psychological distress by 2 years after surgery. This finding indicates that these men have ongoing unmet needs and that more support could be offered where appropriate.

Overall, the 2-year results of this clinical trial are very similar to the 12-week results. Very few differences were detected between Queensland men undergoing robotic-assisted versus open prostate surgery. The robotic-assisted surgery resulted in shorter hospital stays, less pain during recovery and lower blood loss. It also results in less use of tablets and sexual aids, even though erection firmness was considered similar.

Criticisms of the first report

The first publication reporting 12-week results attracted a lot of academic criticism. This came from well-respected people in the field, such as top surgeons and clinical researchers. Some of these criticisms are:

1. Many criticised the first publication for using a 12-week time-point. They state that 12 weeks is not enough time to see the expected differences in side effects and success in treating the cancer. Now we have 2-year results that are mostly consistent with the 12-week results.

2. The two surgeons had different levels of experience. The surgeries in the trial were all done by two surgeons, one performing all the robotic-assisted procedures, and the other all the open surgeries. The robotic-assisted surgeon had trained in the procedure for 2 years. He had done 200 robotic-assisted prostate surgeries at the start of the trial and over 1000 by the end. The surgeon performing the open procedures had 15 years' experience at the start. He had done 1500 procedures at the start of the trial and over 2000 by the end.

The trial was criticised by many for comparing an inexperienced surgeon to an experienced surgeon. These criticisms imply that men having the robotic-assisted surgeries would have had better results if their surgeon had been more experienced. The researchers did not agree with these criticisms, stating in their latest publication that "the quality of the surgical outcomes produced shows that neither of the surgeons was on a novice learning curve or unskilled in their surgical technique".

3. Various parts of the surgeries were performed by trainees. This may have led to differences between the two surgery techniques that were not well-defined.

4. This is a non-blinded trial. It wasn't possible to hide the type of surgery done from the patient, surgeon and others providing card for them. Know what type of surgery was performed may have biased the responses in surveys. For example, the men knowing they had a newer, more sophisticated type of surgery, may have different expectations of success compared to men who knew they had the older style of surgery.

The trial researchers replied that they did not believe that bias was a problem. Their reply stated that: "Although all people involved in assessing the results of this RCT were masked, it was impossible to blind caregivers, patients, and surgeons. In keeping with the professional standing of the medical and nursing staff, all patients received the highest standard of care throughout their management, irrespective of the group to which they were randomised. Of the 308 participants who proceeded to surgery, 252 (151 radical retropubic prostatectomy and 157 robot-assisted laparoscopic prostatectomy)-a creditable 81%-completed their questionnaires at 12 weeks, so we dispute aspersions cast with respect to bias and the validity of findings".

An interesting way to view these results is to consider that a lesser experienced surgeon using the robotic-assisted technique had similar results to a very experienced surgeon using the open surgery. All surgeons need to train for a long time. Every experienced surgeon has trained as an inexperienced surgeon and those operations are performed on real people who have the right to a safe surgical technique. A surgical technique that requires less time to train before proficiency would be an advantage. It's tempting to conclude that this trial showed that robotic-assisted surgery is better for less experienced surgeons. But the trial did not actually test this question. This would need to be tested in a new trial.

What does this all mean for patients?

This landmark Australian clinical trial has demonstrated that for Queensland men having prostate surgery performed by one of two specific surgeons - that there were few detectable differences between robotic-assisted and open surgery. But the relevance of these results for men choosing their surgery today have been questioned on the basis of trial design, size and differences in surgeon experience.

It can be confusing for men considering prostate surgery to hear the there is debate over which is the best surgical technique. Hopefully it is comforting to know that there is an ongoing process by which surgical techniques are assessed and compared. The ultimate aim is to provide the very best in care for patients.

We need to remember that no trial is perfect. While randomised controlled trials are considered the highest level of evidence, they can have results that are inconsistent with what happens in reality. These trials are designed to detect certain levels of difference, depending on how many people are involved. Failure to detect a difference does not mean that none exists, it may be that the number of people in the trial was sufficient for detecting larger differences only. More subtle differences may have gone undetected from smaller trials. In the context of the current trial, this means that differences may still exist between the surgery types - but they did exist, they would not be large differences.

The authors of the trial recommend choosing an experienced surgeon in whom they trust and have rapport, rather than a specific surgery type. But how does a man choose an experienced surgeon? The best place to start is by asking your GP for a recommendation for an experienced surgeon.