Salvage treatment of recurrent, lymph node-positive prostate cancer in the modern era


The advent of imaging tests like the [11C]choline PET/CT scan and others have made it possible to identify, relatively early, the presence of one or more foci of recurrent prostate cancer after first-line therapy, and often these recurrent foci can be found in patients’ pelvic lymph nodes. When one or more positive lymph nodes can be identified in men with recurrent prostate cancer, the next question is, obviously, “What is the best way to treat such a patient?”

Historically, all such patients were treated with some form of androgen deprivation therapy (ADT), but technological advances have now made it possible to investigate other forms of treatment. There are now published data on the use of surgery [i.e., salvage lymph node dissection (sLND)] and targeted radiation therapy [i.e., salvage stereotactic body radiation therapy (sSBRT)] as salvage treatment options for recurrent prostate cancer in one or more isolated positive lymph nodes. However, other forms of treatment may also be possible, including salvage forms of high-intensity focused ultrasound (HIFU) and focal laser therapy to name just two. We expect to see more research in this area in the future.

To give a quick introduction to the available literature on this topic it is worth mentioning three papers.

As long ago as 2011, Rigatti et al. reported data from a series of 72 patients with biochemical recurrence after a radical prostatectomy, all of whom were identified as having positive lymph nodes by an [11C]choline PET/CT scan. All 72 patients were treated by sLND. In this cohort:

  • Mean and median follow-up after sLND were 39.4 and 39.8 months, respectively.
  • 41/72 patients (56.9 percent) demonstrated a biochemical response to treatment (a PSA level of < 0.2 ng/ml at 40 days after sLND).
  • Actuarial 5-year survival rates were
    • 19 percent for biochemical recurrence-free survival
    • 34 percent for clinical recurrence-free survival
    • 75 percent for prostate cancer-specific survival

In this early report, Rigatti et al. concluded that

Salvage LND is feasible in patients with [biochemical recurrence] after [radical prostatectomy] and nodal pathologic uptake at [11C]choline PET/CT scan. Biochemical response after surgery can be achieved in a consistent proportion of patients. Although most patients invariably progressed to [biochemical recurrence] after surgery at longer follow-up, 35 percent of patients showed the absence of [clinical recurrence] at 5 [years].

The Mayo Clinic research team (Karnes et al.) have recently reported data from a series of 52 patients, all of whom received sLND for nodal recurrence of prostate cancer detected by [11C]choline PET/CT scans after radical prostatectomy, and all of whom were treated between 2009 and 2013. This is the largest series of such patients treated in the US to date.

  • 41/52 patients (78.8 percent) had received some form of second line therapy after radical prostatectomy and before sLND.
  • Median patient age at sLND was 60 years.
  • Median PSA level was 2.2 ng/ml.
  • The median number of lymph nodes dissected was 21.5.
  • The median number of positive nodes was 3.5.
  • Following sLND
    • 24/52 patients (46.2 percent) required no further treatment.
    • 18/52 patients (34.6 percent) received ADT.
    • 10/52 patients (19.2 percent)received multiple different treatments.
  • At last follow-up (at a median of 20 months post-sLND)
    • 30/52 patients (57.7 percenthad a PSA level of < 0.2 ng/ml.
    • 39/52 patients (75.0 percent) were free of systemic progression.
    • Overall survival was 96.2 percent.
    • Two patients had died of prostate cancer.
  • Actuarial 3-year survival rates were
    • 45.5 percent for biochemical recurrence-free survival
    • 46.9 percent for systemic progression-free survival
    • 92.5 percent for prostate cancer-specific survival

Karnes et al. concluded that, “Although followup was short and the study lacked a randomized control group, salvage lymph node dissection may represent a valid treatment option” for patients with a rising PSA and imaging-identified positive lymph nodes.

Another Italian research group (Ponti et al.) have evaluated the efficacy of sSBRT in the management of a small series of 16 men with recurrent prostate cancer and 18 isolated lymph nodes treated between 2008 and 2013. All patients had been given an [11C]choline PET/CT scan prior to sSBRT (12 to 35 Gy, delivered in up to five daily fractions). They report that:

  • 10/16 patients received ADT in conjunction with the sSBRT.
  • Average follow-up was a mean of 29.35 months (median, 29.38 months; range, 6.3 to 68.8 months).
  • 15/16 patients (94 percent) exhibited local disease control and a decrease in serum PSA levels.
  • 1/16 patients (6 percent) had an in-field progression.
  • For the six patients who received no ADT at the time of SBRT, mean time to initiation of palliative ADT was 23.7 months (range, 2.5 to 51 months).
  • At last follow-up,
    • 8/16 patients (50 percent) had active prostate cancer disease.
    • Average (mean) time to biochemical progression after completion of sSBRT was 7.9 months.
    • The 2-year biochemical relapse-free survival was 44 percent.
    • 1/16 patients (6 percent) died of prostate cancer.
    • Overall survival was 94 percent.
    • Late toxicity was observed in 1 patient who had  Grade 3 gastrointestinal toxicity.

Ponti et al. conclude that:

SBRT seems to be safe, effective, and minimally invasive in the eradication of limited nodal recurrence from oligometastatic prostate cancer. SBRT is well tolerated by patients with low toxicity and yielded a local control of the disease.

What is evident from these data is that at least some patients with recurrent prostate cancer after radical prostatectomy and evidence of isolated nodal metastasis (identifiable by [11C]choline PET/CT scanning techniques) can indeed be treated rather successfully by either sLND or sSBRT. What is less clear is the ability to predict who will respond well to such therapy and who will not, although the usual risk factors (higher PSA levels, higher Gleason scores, and a higher number of positive lymph nodes) are clearly indicative of a lower probability of a good outcome in the long-term.

10 Responses

  1. Is the above mentioned sLND surgery an open surgery or is it a robotic surgery?

  2. sLND can be performed using either open or laparoscopic, robot-assisted procedures. The precise methodology would probably depend on the precise location of the lymph node or nodes to be dissected.

  3. I think it’s worth adding a point about the difficulty of salvage lymph node treatment.

    My understanding is that sLND is very difficult. The lymph nodes are practically invisible against surrounding tissue and difficult to locate. Especially in the pelvic area, there is movement, and a LN identified on an earlier CT scan may be in a different position. There has been some progress using fluorescent dyes to better see them. Several studies suggest that extensive pelvic lymph node dissection (ePLND) has better oncologic outcomes than PLND when done at the same time as RP. However, extensive LN removal carries risks of lymphocele and lymphedema.

    In the Ponti et al. study, there were 18 positive nodes in 16 patients, so most of their well-selected sample had just one positive node treated. There may be value in treating one or two, perhaps up to five LNs with highly targeted sSBRT. But the problem arises that as more LNs are infected, the probability that the cancer has disseminated broadly throughout the lymphatic system goes up geometrically. Remember that LNs are connected in networks, rather than serially. So, it’s seldom going to be curative to treat just one or two. Radiation oncologists may choose to treat an area of several inches around pelvic LNs identified by the PET scan, or they may choose to treat the entire area, recognizing the potential for bowel toxicity in doing so.

    The Sitemaster may know of cases of LNs treated with ablation, but I have never heard of that, and have serious doubts about its potential.

    I very much agree that in certain well-selected individuals salvage LN treatment may be curative, but it’s very difficult given our present knowledge and technology to sort out those who can benefit from those who can’t.

  4. Thank you both, Allen and Sitemaster, for this post and your comments — it has been helpful and very timely working with patients/caregivers seeking current advice, and I have forwarded. Unfortunately only the abstracts are available without subscription and there are several open questions.

    While a 40% success rate is evident, there is a lot of information lacking — for example the period between identifying recurrent disease and submitting for surgery or radiation. The abstracts do not indicate whether the men who nailed the disease with a lymphadendectomy had a shorter time period between recurrence and surgery than those who did not. Nor is there information about timing in the radiation abstract.

    Bottom line is that debulking lymph nodes has some credibility and appears to merit consideration; however, (1) a lot more research and information is needed before we can rely on a 40% success rate, and (2) as correctly pointed out by Allen and Mike in the comments, the risk of surgery must be considered — especially if the nodes are inaccessible.

  5. The median PSA was 2.2 in the above article at the time of SLND surgery. Not being very good at math, would anyone take a guess at the highest PSA at time of this surgery?

    Thank you!

  6. Dear Seabright:

    The so-called “interquartile range” or IQR was from 1.4 to 3.7 ng/ml, which implies (I think) that almost certainly none of these patients had a PSA level of much more than 4 ng/ml by the time they were given SLND. But I am also no statistician, so someone else may be able to give you a better answer.

  7. The study gives the interquartile range as 1.4-3.7 ng/ml. That’s the range that excludes the lowest quarter of PSAs and the highest quarter of PSAs, so it represents the middle half of all patients. They give this statistic to exclude the extremely low and high PSAs that would otherwise skew the results. You can’t just guess at the highest PSA because the PSAs were not normally distributed — they are skewed to the right (high PSA) with more extremely higher values than there are lower values.

  8. Thanks Mike!

  9. See … I said I wasn’t a statistician! :O)

  10. Thanks Allen for the explanation!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: