How much prednisone is needed with abiraterone?

Abiraterone (Zytiga) inhibits the enzymes necessary for the adrenal glands to produce androgens. The adrenals are the secondary source of androgens after the testicles. Unfortunately, the same enzyme is needed for the adrenals to produce cortisol, so a glucocorticosteroid (in the form of prednisone, dexamethasone, or methylprednisolone) has to be replaced when taking abiraterone. When abiraterone is prescribed for men with metastatic hormone-sensitive prostate cancer, 5 mg once daily (5 mg qd) or sometimes 2.5 mg twice daily (2.5 mg bid) is taken for replacement. When abiraterone is prescribed for men with metastatic castration-resistant prostate cancer, 5 mg twice daily (5 mg bid) is taken with it, because the anticipated duration of taking the steroids is shorter. Recently it was found that switching from prednisone to dexamethasone 0.5 mg once a day (0.5 mg qd) can extend the duration of abiraterone effectiveness (see this link).

The biggest danger of taking too little corticosteroid to replace what is lost is a condition called a "syndrome of secondary mineralocorticoid excess." This occurs because the pituitary gland reacts to the lack of cortisol by producing a hormone called ACTH (this is called "negative feedback"). ACTH increases the production of mineralocorticoids (like aldosterone), hormones that increase blood pressure, lower potassium and cause edema in the limbs.

Cortisol has many functions, including energy production, control of inflammatory response (e.g., preventing arthritis), and preventing allergies and runaway immune response. It also has mental effects, affecting mood and memory formation. Symptoms of too little cortisol may include fatigue, dizziness (especially upon standing), nausea, fever, weight loss, muscle weakness, joint pain, mood changes, and the darkening of regions of the skin.

The danger of taking too much corticosteroid may include insulin resistance, a decrease in lean body mass, increase in fat accumulation, and decrease in bone mineral density. It is also immunosuppressive, and may cause tissue breakdown (catabolism) and gastritis. Adverse effects increase with dose and duration of use.

Glucocorticosteroids have been found to have independent anti-cancer activity (see this link). The effect is short-lived as resistance eventually arises. It is also given to mitigate some of the side effects of chemotherapy like emesis/nausea and peripheral edema.

Attard et al. conducted a randomized clinical trial among men taking 1000 mg/day abiraterone for metastatic castration-resistant prostate cancer at 22 hospitals in 5 countries in 2013-2014.

• 41 received it with prednisone 5mg bid (P5 bid)
• 41 received it with prednisone 5 mg qd (P5 qd)
• 40 received it with prednisone 2.5 mg bid (P2.5 bid)
• 42 received it with dexamethasone 0.5 mg qd (D0.5 qd)

The primary outcome measured was mineralocorticoid excess through 24 weeks of treatment as indicated by elevated blood pressure or a blood test for low potassium (hypokalemia). They also measured serum levels of ACTH, which gets elevated if there is not enough glucocorticoid. For side effects of too much glucocorticoid, they measured insulin resistance, loss of lean body mass, gain of body fat, and loss of bone mineral density. For the benefits, they measured suppression of androgen precursors, the % of patients in whom PSA declined by at least 50%, the duration of radiographic progression-free survival, and the patient-reported change in quality of life.

	P5 bid	P5 qd	P2.5 bid	D0.5 qd
Mineralocorticoid excess	29%	63%	40%	30%
(95% confidence range)	17%-46%	47%-77%	26%-56%	17%-46%
Grade 3 hypertension	7%	22%	13%	7%
Grade 3 hypokalemia	0%	7%	0%	0%
Change in ACTH (pmol/L)	-1.1	9.0	4.0	-1.8
Change in fasting serum insulin (insulin resistance)	Not statistically significant	Not statistically significant	Not statistically significant	significant
Change in lean body mass	-6%	-3%	-6%	Not statistically significant
Change in total body fat	12%	Not statistically significant	Not statistically significant	19%
Change in bone mineral density	Not statistically significant	Not statistically significant	Not statistically significant	-2%
Androgen precursor suppression	-81%	Not statistically significant	Not statistically significant	-88%
PSA declined by ≥ 50%	63%	78%	60%	88%
Radiographic Progression-free survival	18.5 months	15.3 months	12.8 months	26.6 months
Quality of Life change	Not statistically significant	Not statistically significant	Not statistically significant	Not statistically significant

Although sample sizes were not large enough to directly compare the treatments, the data suggest that P5 bid and D0.5 qd do a good job of preventing mineralocorticoid excess, whereas P5 qd does not. P5 bid and D0.5 qd seem to cause body changes. D0.5 qd seems to have superior oncological effectiveness.

There was tremendous individual variation. It seems prudent to start with the prescribed dose (P5 bid) and monitor body changes, or to start at the lower dose (P5 qd) and to monitor blood pressure and potassium levels.

In a retrospective study, Gill et al. found that castration-resistant men at Huntsman Cancer Center who refused prednisone suffered no worse from the syndrome of mineralocorticoid excess when given 50 mg/day Inspra (eplerenone - an aldosterone antagonist) compared to men who used 10 mg/day prednisone. They also lost weight, perhaps water-weight. Patients taking 10 mg/day of prednisone may wish to discuss this alternative.

Evidence for Dose Escalation in Adjuvant/Salvage Radiation

It is well known that prostate cancer is relatively radio-resistant compared to other kinds of cancer. While dose escalation (most recently by increasing the biologically effective dose using hypofractionated dose (more that 2.0 Gy per session) delivery or brachytherapy boost therapy) has become the mainstay in primary radiation therapy, doses delivered for adjuvant or salvage radiation has stayed about 10 Gy lower. Recently, Dr. King's analysis of the dose responsiveness of salvage radiation questioned this supposition (see this link). While his mathematical arguments provide us with intriguing plausibility, only clinical evidence from a randomized clinical trial can change practice.

We now have Level 1 evidence that expanding the adjuvant/salvage treatment field to include the pelvic lymph nodes improves the oncological outcomes in men with higher PSA at the time of salvage radiation.

Link et al. conducted a small, retrospective study among 120 locally advanced (stage T3/4) post-prostatectomy patients at the University of Heidelberg between 2009 and 2017. All were lymph node negative.

• 43 received whole pelvic radiation therapy (WPRT)- 62% received 79.3 Gy to the prostate
• 77 received radiation to the prostate bed only (PBO)- 70% received 79.3 Gy to the prostate
• Biologically equivalent dose (2 Gy) to the prostate was 79.3 Gy ("high dose") if they had positive margins or PET/CT/MRI imaging-detectable prostate bed tumors (62% of patients), 71.4 Gy ("low dose") if they had negative margins (38% of patients).

Median freedom from biochemical failure was:

• longer among those who got the higher dose: 76 months vs 21 months
• longer among those who received WPRT vs PBO: 68 months vs 32 months

There is a lot of overlap in treatments, so it is impossible to tease out the effect that each had on the oncological outcomes. Almost all of those who received the escalated dose also had positive margins - a known factor for predicting success of adjuvant/salvage radiation. Also, almost all men who had adjuvant radiation had positive margins and dose escalation - adjuvant radiation has proven to be more successful than "wait-and-see" in 3 major randomized clinical trials.

Toxicity increased with both dose and size of the treatment field. Grade ≥ 2 toxicity was reported by:

• 3.4% among those who received low dose and PBO
• 12.5% among those who received high dose and PBO
• 15.4% among those who received low dose and WPRT
• 36.7% among those who received high dose and WPRT
• No reports of Grade 3 gastrointestinal toxicity
• 13% Grade 3 urinary toxicity among high dose patients, none among low-dose patients

This is a far cry from the randomized clinical trial we need for practice-changing dose escalation for adjuvant/salvage radiation. However, we can't rule out that there is no oncological benefit to dose escalation. It remains unknown what proportion of these high-risk patients would have done just as well with lower doses and smaller treatment fields. The increase in toxicity with dose and treatment field means that patients ought not jump into this without understanding the risks and discussing them with their radiation oncologists.

"If all you have is a hammer, everything looks like a nail"

It was called "instrument bias" by Abraham Maslow and Abraham Kaplan, but for present purposes, we'll call it "specialty bias" - over-reliance on the tool one is most familiar with. Kishan et al. conducted a survey among urologists ("Uros") and radiation oncologists ("ROs") concerning their opinions about how best to treat high-risk patients in various situations from initial treatment to recurrence after initial treatment. They tabulated responses from:

• 846 ROs and 407 Uros
• 63% of ROs and 96% of Uros practiced in the US; the rest mainly in Australia and NZ
• They had a median of over 10 years of experience
• 41% of ROs and 51% of Uros were in private practice

Initial Treatment of High Risk Patients

ROs were 5 times more likely to believe that initial treatment with radiotherapy (RT) with androgen deprivation therapy (ADT) and with local salvage therapy, if needed, was preferred. They were also twice as likely to believe that it offered the patient equivalent outcomes as radical prostatectomy (RP) and salvage radiation (SRT), if needed. Uros were 4 times more likely to believe that RP ± SRT was the preferred treatment. Only 29% believed that RT had a place in initial treatment.

Preferred Initial Treatment	RO	Uro
RP±SRT	18%	71%
RT+ADT±local salvage	30%	6%
No preference	52%	23%

The "right" answer is...

Currently, the American Urological Association (AUA) and the American Society of Radiation Oncologists guidelines state:

"Clinicians should recommend radical prostatectomy or radiotherapy plus ADT as standard treatment options for patients with high-risk localized prostate cancer. (Strong Recommendation; Evidence Level: Grade A)"

NCCN guidelines also list both as options. They note, however:

"A large, multicenter, retrospective cohort analysis that included 1809 men with Gleason score 9–10 prostate cancer found that multimodality therapy with EBRT, brachytherapy, and ADT was associated with improved prostate cancer-specific mortality and longer time to distant metastasis than either radical prostatectomy or EBRT with ADT. In addition, an analysis of outcomes of almost 43,000 men with high-risk prostate cancer in the National Cancer Database found that mortality was similar in men treated with EBRT, brachytherapy, and ADT versus those treated with radical prostatectomy, but was worse in those treated with EBRT and ADT . (MS25)

This can only be decided definitively by a randomized clinical trial, but given the difficulties of recruiting for such a trial, patients must make the decision based on lower level evidence.

Adjuvant RT (ART) and SRT for High-Risk Patients after RP

ROs were 2.7 times as likely to advocate for ART with undetectable PSA, and were twice as likely to believe that ART is underutilized. Conversely, Uros were 2.5 times more likely to approve of waiting until PSA has risen to 0.2 ng/ml. Most believed it is utilized appropriately (it is seldom utilized in this situation) or overutilized. About 2 in 5 ROs and Uros were OK with early SRT.

Two-thirds of ROs thought SRT was appropriate with 2 consecutive rises, or at any detectable level or any level under or equal to 0.1. Less than half of Uros held that belief. ROs were more likely than Uros to believe that SRT is underutilized.

ART Appropriate	RO	Uro
With undetectable PSA	43%	16%
Early SRT with detectable PSA < 0.2 ng/ml	42%	43%
Delayed SRT with PSA ≥ 0.2 ng/ml	16%	41%

ART Under/Over-Utilized	RO	Uro
Underutilized	75%	38%
Appropriately utilized	21%	43%
Overutilized	4%	19%

Lowest PSA at which SRT is Appropriate	RO	Uro
Detectable	15%	7%
≤ 0.1	21%	19%
At least 0.2	29%	35%
At least 0.4	4%	12%
At least 1.0	1%	7%
At least 2 consecutive rises	30%	20%

SRT Under/Over-Utilized	RO	Uro
Underutilized	65%	43%
Appropriately utilized	34%	52%
Overutilized	1%	5%

The "right" answer is...

The 2019 AUA/ASTRO guidelines state:

Guideline Statement 2. Patients with adverse pathologic findings including seminal vesicle invasion, positive surgical margins, and extraprostatic extension should be informed that adjuvant radiotherapy, compared to radical prostatectomy only, reduces the risk of biochemical recurrence, local recurrence, and clinical progression of cancer. They should also be informed that the impact of adjuvant radiotherapy on subsequent metastases and overall survival is less clear; one of three randomized controlled trials that addressed these outcomes indicated a benefit but the other two trials did not demonstrate a benefit. However, these two trials were not designed to identify a significant reduction in metastasis or death with adjuvant radiotherapy. (Clinical Principle)

 Guideline Statement 3. Physicians should offer adjuvant radiotherapy to patients with adverse pathologic findings at prostatectomy including seminal vesicle invasion, positive surgical margins, or extraprostatic extension because of demonstrated reductions in biochemical recurrence, local recurrence, and clinical progression. (Standard; Evidence Strength: Grade A)

Guideline Statement 8. Patients should be informed that the effectiveness of radiotherapy for PSA recurrence is greatest when given at lower levels of PSA.  (Clinical Principle) 

"...patients should be advised that if recurrence is detected without evidence of distant metastases, then RT should be administered at the earliest sign of PSA recurrence ..."

The 2019 NCCN guidelines state:

Indications for salvage RT include an undetectable PSA that becomes subsequently detectable and increases on 2 measurements or a PSA that remains persistently detectable after RP. Treatment is more effective when pre-treatment PSA is low and PSADT is long.

Salvage for High-Risk Patients after RT

About 2/3 of Uros believed that RP and Cryo are appropriate salvage therapies after biochemical recurrence following primary RT (when imaging was negative for distant metastases). Very few thought salvage radiation or other ablation therapies were appropriate. Among ROs, most approved of salvage RP, but sizeable minorities felt that salvage brachytherapy and salvage cryo was appropriate. About 1 in 5 thought salvage SBRT or salvage HIFU was appropriate.

Salvage therapies after RT failure	RO	Uro
RP	56%	66%
LDR brachytherapy	34%	6%
HDR brachytherapy	42%	10%
SBRT	20%	5%
Cryoablation	40%	69%
HIFU	21%	13%

However, 2/3 of both ROs and Uros and agreed that salvage after RT was not as effective as salvage after RP.

Is salvage after RT as effective as SRT after RP?	RO	Uro
No	64%	66%
Yes	36%	34%

And most of both groups believe that focal salvage is inferior to whole gland salvage after RT failure. But Uros were much more likely to hold this belief vs ROs.

Focal salvage=whole gland salvage?	RO	Uro
No	56%	75%
Yes	44%	25%

The "right" answer is...

NCCN guidelines state:

Salvage RP is an option for highly selected patients with local recurrence after EBRT, brachytherapy, or cryotherapy in the absence of metastases, but the morbidity (ie, incontinence, loss of erection, anastomotic stricture) is high and the operation should be performed by surgeons who are experienced with salvage RP. 

Brachytherapy can be considered in men with biochemical recurrence after EBRT. In a retrospective study of 24 men who had EBRT as primary therapy and permanent brachytherapy after biochemical recurrence, the cancer-free and biochemical relapse-free survival rates were 96% and 88%, respectively, after a median follow-up of 30 months.Results of a phase 2 study of salvage HDR brachytherapy after EBRT included relapse-free survival, distant metastases-free survival, and cause-specific survival rates of 68.5%, 81.5%, and 90.3%, respectively, at 5 years.

HIFU also has been studied for treatment of radiation recurrence. Analysis of a prospective registry of men treated with HIFU for radiation recurrence revealed median biochemical recurrence-free survival at 63 months, 5-year OS of 88%, and cancer-specific survival of 94%.

For a discussion of salvage therapies after RT, see this link. Salvage after RT is generally as effective or better than SRT after RP. Focal salvage is as effective as whole gland salvage in many cases.

What to do about specialty bias

It is not surprising that specialists are more knowledgeable about, and more favorably disposed to their own field (otherwise, they're in the wrong job). However, there are knowledge gaps even within their own field. Many patients expect their doctors to be knowledgeable about their own field, and to be able to compare it to other therapies. We have seen that this is an unrealistic expectation. Here are some recommendations for patients:

• Don't expect a specialist to be knowledgeable outside his own field. Consult with a variety of specialists, if possible.
• Remember that it is radiation oncologists, not urologists, who treat and follow-up on SRT for surgical recurrences. ROs are more experienced than Uros at SRT.
• Salvage brachytherapy after RT recurrence is highly specialized - there aren't many practitioners with expertise. Salvage ablation and salvage SBRT specialists are even more rare. Patients should avoid salvage surgery due to its high morbidity.
• Second opinions are critical. Finding specialists may involve travel, which may be precluded by cost/insurance limitations.
• For patients who are inclined to research these topics themselves, they must be aware that the quality of research on all of these topics is low to moderate. There have been very few randomized clinical trials focussed on high risk patients, so we have to make informed judgments based on observational studies or single-arm, single-institution clinical trials in most situations. AUA, ASTRO and NCCN guidelines are updated and may be their best source of information (other than this blog 😉).

Focussing on physicians, the authors recommend interdisciplinary clinics, which might work well in academic centers. However, a substantial proportion of both ROs and Uros were in private practice. They also suggest continuing medical education, and cross-specialty training. Professional organizations typically do not currently require this. Uros do not typically attend ASTRO meetings, nor do ROs attend AUA meetings.

- with thanks to Dr. King for allowing me to see the full text and questionnaire.