How can hot flushes in men being treated for prostate cancer be managed?

Download 192.5 Kb.
Size192.5 Kb.

Medicines Q&As

Q&A 383.4
How can hot flushes in men being treated for prostate cancer be managed?

Prepared by UK Medicines Information (UKMi) pharmacists for NHS healthcare professionals

Before using this Q&A, read the disclaimer at

Date prepared: September 2015


Hormonal manipulation methods are used in the management of prostate cancer. This includes methods such as androgen withdrawal, also known as androgen deprivation therapy (ADT), using luteinising hormone-releasing hormone agonists goserelin and leuprolide or androgen receptor blockade by using anti-androgens cyproterone and bicalutamide.(1;2) The withdrawal of the sex hormones not only causes disruption of the thermoregulatory centre in the hypothalamus but also causes noradrenaline fluctuations, resulting in hot flushes (also called vasomotor symptoms or hot flashes).(3-5) The exact mechanism of hot flushes is not fully understood.(6) The thermoregulatory centre in the hypothalamus maintains the core body temperature within a normal thermoregulatory zone and it is thought that a dysfunction of this centre as well as a downward shift and narrowing of the thermoregulatory zone may be the cause.(6) In women, decreased endogenous oestrogen concentrations causes these changes and it is thought that decreased testosterone in men has a similar effect.(7)

The incidence of hot flushes in men undergoing ADT has been estimated at 55-85% and manifest as a sensation of increased upper body temperature, skin reddening and sometimes profuse sweating, followed by a cold, clammy feeling.(4;5) The intensity of the hot flush can be described as

  • Mild: generalised warmth or flushing lasting less than 3 minutes

  • Moderate: more warmth and/or flushing than a mild flush, lasting <5 minutes

  • Severe: hot or very hot, accompanied by anxiety or irritability, heavy perspiration, extreme discomfort and may require a change of clothing, lasting <10 minutes

  • Very severe: very hot sensation lasting <30 minutes, drenching perspiration requiring a change of clothing or bed and a bath/shower.(8)

Hot flushes are not necessarily self-limiting, can be incapacitating if they are severe and can last long after therapy has been discontinued.(2;4) Hot flushes occurring at night can lead to frequent awakening, which in the long-term can cause chronic fatigue, irritability and altered memory and attention span.(9)

Unlike the use of oestrogen replacement therapy for treating hot flushes in postmenopausal women, androgen replacement therapy cannot be used because the treatment aim of prostate cancer is to deprive cancer cells of androgens.(10) NICE clinical guidelines for treatment of prostate cancer recommend that medroxyprogesterone (20 mg per day) is offered initially for 10 weeks to manage hot flushes and treatment effect evaluated at the end of the 10 week period.(11) If medroxyprogesterone is ineffective or not tolerated NICE advises to consider cyproterone acetate (see below for dosing) or megestrol acetate (20mg twice a day for 4 weeks). (11) Note, however that only a 160mg megestrol tablet is available in the UK. (12)
Cyproterone acetate has progestogenic activity, exerting a negative feedback on hypothalamic receptors and therefore reducing both gonadotrophin release and testicular androgens.(13) It is licensed to treat hot flushes in patients undergoing LHRH analogue treatment or orchiectomies. The cyproterone dose for suppression of "flare" with initial LHRH analogue therapy is 100mg BD alone for 5 - 7 days, followed by 100mg BD for 3 – 4 weeks together with the LHRH analogue therapy. For the treatment of hot flushes in patients under treatment with LHRH analogues or who have had orchidectomy a 50 mg starting dose, with upward titration if necessary within the range 50-150mg/day, is recommended, in divided doses.(13)
Medroxyprogesterone and megestrol acetate both do not have a UK marketing authorisation for this indication. The prescriber should therefore follow relevant prescribing guidance, taking full responsibility for the recommendations. What evidence is there to support these, and other off-label treatment for managing hot flushes in men with prostate cancer?
NICE recommends that medroxyprogesterone, megestrol acetate or cyproterone are used to control hot flushes caused by long-term androgen suppression (11) but other therapies have been used.
It has been postulated that neurotransmitters such as noradrenaline, serotonin and endorphins, regulate the thermoregulatory centre in the hypothalamus, so agents which inhibit reuptake of these neurotransmitters could have beneficial effects on hot flushes.(10) A number of different treatments for hot flushes in men treated for prostate cancer have been studied. Hot flush diaries were used in the studies to assess severity and frequency of the hot flushes and patient’s mood and quality of life. Many of the studies are short, do not carry the strength of a longer, prospective, placebo-controlled trial and may prevent firm conclusions being drawn. Five short, randomised controlled trials have been carried out involving the use of transdermal clonidine, gabapentin, venlafaxine, MPA and cyproterone. The studies are summarised below with further details in table 1.

Small studies have evaluated the efficacy of transdermal or oral clonidine for reducing hot flushes, with varied results.(14-16) The smaller studies showed a positive effect of clonidine on the frequency and duration of hot flushes,(14;16) while the larger study did not.(15) Note that clonidine patches are not licensed in the UK.


Gabapentin 900mg/day (in divided doses) has been shown to be significantly more effective in treating hot flushes than 300mg/day (17) in a phase 3 RCT, but when treatment was continued in an open-label study, a 600mg/day dose was favoured, possibly because of twice, rather than three times a day, dosing.(18) No significant difference between the combined gabapentin groups and placebo was seen but a dose-response was clear, with higher doses producing more positive results.

Medroxyprogesterone acetate (MPA)]

Intramuscular injections of MPA have been successfully used to reduce the frequency and severity of hot flushes.(19;20) A 400mg or 150mg depot injection was used in the studies (19;20); the two IM preparations available in the UK are the 150mg Depo-Provera, which is given every 12 weeks or the 104mg SAYANA PRESS, given every 13 weeks for contraception.(12) Medroxyprogesterone acetate 20mg orally has been used in one comparative randomised study and is recommended by NICE. (11;21)

Megestrol acetate

Megestrol has been shown to be an effective treatment in reducing the frequency and severity of hot flushes, in a short study (22) with a longer follow-up.(23) Note that only a 160mg megestrol tablet is available in the UK (12) but doses used for treating hot flushes in the studies ranged from <20mg to 160mg/day.(22;23)


A small randomised placebo controlled study concluded that paroxetine 10mg reduced the frequency and severity of hot flushes in prostate cancer patients over an 8 week period.(24) Two small, uncontrolled, open-label studies have evaluated the efficacy of paroxetine 10mg-37.5mg in men receiving ADT and experiencing 7-14 hot flushes a week for at least 1 month.(10;25) The studies showed that paroxetine reduced the frequency and severity of hot flushes over a 4-week treatment period and also improved quality of life (QoL) measures.


Case reports of five men having ADT for prostate cancer and who were treated with sertraline for anxiety or depression, showed that sertraline also had a positive effect on the hot flushes that the men were experiencing.(26) Up to 10 hot flushes a day were occurring, some of which were accompanied by drenching sweats, as well as several hot flushes during the night. When sertraline doses were titrated up to 75-100mg/day the men noticed a considerable improvement in the hot flushes, in frequency, intensity and duration.


Venlafaxine for treating hot flushes has been evaluated in two small open-label studies and two randomised controlled trials.(21;27-29) Men having ADT for at least 6 weeks prior to study entry and experiencing at least 14 hot flushes a week for at least one month were enrolled. After four weeks of venlafaxine 12.5mg bd treatment, an improvement in the number and severity of hot flushes was seen.

However, a short randomised, double-blind study in 309 men having ADT, showed that cyproterone 100mg daily and MPA 20mg/day were statistically significantly more effective than venlafaxine 75mg/day in reducing the number and severity of hot flushes.(21) Significantly more patients considered treatment with cyproterone or MPA to have good efficacy than those who considered venlafaxine to have good efficacy. No significant difference in the number of adverse events between the three treatments was seen. Another randomised, double-blind placebo controlled study assessed venlafaxine, soya protein or in combination to treat hot flushes in 120 men with prostate cancer. However the trial was stopped early due to a lack of effect in reducing hot flushes with all trialled treatments.(27)
Table 1: Study reports

Limitations of many of these studies include short treatment durations and open-label designs.


Study design and treatment




  • Open-label, pilot study.

  • Clonidine patches, 0.1mg/day, changed weekly, in 7 men experiencing post-orchiectomy hot flushes at night and between <5 to >10 during the day, lasting a median 3.5 minutes.

  • The minimum follow-up was 7 months. All patients experienced some relief from the transdermal clonidine, with better results in those patients suffering with fewer hot flushes at baseline. Symptoms resolved fully in two patients who experienced up to 5 flushes per day but only improved in those experiencing >10. The main side effect was drowsiness.

  • A small, open-label, pilot study. The very nature of an open-label study can lead to bias and results should be interpreted cautiously.


  • Open-label, pilot study.

  • Transdermal clonidine (0.1-0.6mg/day, n=2) or oral clonidine (0.1-0.6mg/day, n=2) in men receiving ADT and experiencing hot flushes

  • Treatment was given for an average of 9 months (range 3-24).

  • All patients had a partial response to clonidine within 2 weeks, with a reduction in frequency of the hot flushes, and, in the men treated with transdermal clonidine, a reduction in the duration, from 10 minutes to 1-2 minutes.

  • Small, open-label, pilot study, with data presented per individual, rather than as a group.


  • Double-blind, randomised, cross-over trial.

  • Weekly transdermal clonidine (equivalent to a daily oral dose of 0.1mg) or placebo, given for 4 weeks, total trial duration 9 weeks.

  • 78 men who had undergone orchiectomy (medical, 26% or surgical, 74%) and who were suffering with at least seven hot flushes a week for a month.

  • Fifty men completed the 9 weeks.

  • The median frequency of hot flushes during baseline was 8 per day, with a median severity of 1.7 (1 = mild, 4 = very severe). At the end of each treatment period, a minor trend for lower hot flush frequency was seen with clonidine but no significant advantage over placebo was shown.

  • When asked which patch worked best to reduce the hot flushes, the majority of patients (47%) could not tell which patch was better, 34% preferred clonidine and 19% preferred placebo.

  • Treatment with clonidine was associated with a higher incidence of dry mouth and redness under the patch but most patients could not tell which treatment period was better in terms of adverse events.

  • The negative result was not due to insufficient sample size; the study had at least 85% power to detect a difference of 25% or more in the value of the hot flush frequency ratio or hot flush score ratio between treatment arms.


  • Randomised, double-blind, placebo-controlled phase 3 study.

  • Gabapentin 300mg for 28 days (n=54), gabapentin 300mg daily for 7 days then 300mg bd for 21 days (n=53), gabapentin 300mg for 7 days, then 300mg bd for 7 days, then 300mg tds for 14 days (n=54), or placebo mirroring each gabapentin dosing schedule (n=53).

  • Men undergoing ADT who had at least 14 hot flushes a week for at least 1 month.

  • Fifty patients per group would provide 80% power to detect a clinically meaningful difference in the primary end point between gabapentin and placebo.

  • No significant differences were seen between the collective gabapentin groups and the placebo group for the median changes from baseline in:

  • Hot flush score (primary endpoint): -3.3 (gabapentin) vs. -2.8 (placebo), difference = -0.6, [95% CI -2.3 to 1.0, p=0.48].

  • Per treatment median changes were -2.4 (median -29.7% decrease from baseline, 300mg), -3.1 (-33.8%, 600mg) and -4.3 (-44.4%, 900mg, p=0.05 vs. 300mg).

  • Mean hot flush scores fell by -4.1 (placebo), and by -3.2, -4.6 and -7.0 for gabapentin 300mg, 600mg and 900mg respectively.(18)

  • Hot flush frequency: -2.1 (gabapentin) vs. -1.6 (placebo), difference = -0.6, [95% CI -1.6 to 0.3, p=0.19].

  • Per treatment median changes from baseline (median % decrease) were -1.8 (-22.8%, gabapentin 300mg), -2.0 (-31.8%, 600mg) and -2.6 (-45.5%, 900mg, p=0.02 vs. placebo and p=0.03 vs. 300mg).

  • Gabapentin had a significant positive effect on hot flush control (p=0.03 vs. placebo) and how they affected quality of life (p=0.01 vs. placebo), mainly driven by significant differences between the gabapentin 900mg/day group vs. placebo.

  • A higher incidence of appetite loss and constipation occurred in the placebo group.

  • One of the better designed and reported studies.

  • Although no significant difference between the combined gabapentin treatment and the placebo treatment was seen, a dose-response effect was clear, with higher doses producing more positive results.


  • 8-week, open-label, uncontrolled, continuation phase of the study above (n=147) (13).

  • Patients treated with placebo were started on gabapentin 300mg/day; all could increase or decrease their dose by 300mg/day each week according to control or toxicity, with a maximum dose of 900mg/day.

  • Mean dose taken was 600mg/day (300mg bd).

  • By the end of the continuation phase, all patients had achieved similar reductions in hot flush frequency and scores.

  • From the original baseline to week 12, mean hot flush scores fell by 44%, 46%, 48% and 49% in the original placebo, 300mg, 600mg and 900mg groups. Corresponding reductions in hot flush frequency were 49%, 57%, 51% and 51%.

  • No increase in adverse effects was seen.

  • Significant improvements in two QoL measures were seen in both the placebo and 300mg groups: ‘how distressing are your hot flushes’ and ‘how satisfied are you with your hot flush control’. Patients in the original placebo group showed a significant improvement in the ‘how much did the hot flushes affect your quality of life’ measure, in the continuation phase.

  • An open-label continuation phase.

  • The majority of patients took 600mg/day, suggested that three-times a day dosing may be inconvenient.


  • Open-label, uncontrolled study.

  • Medroxyprogesterone acetate (MPA) 400mg by intramuscular (IM) injection in 55 men on ADT and experiencing hot flushes.

  • The men were followed for a mean of 17.2 months (range 6-36 months) and significant improvements in symptoms were seen in 51 men (92.7%).

  • A single dose was administered to 27 men, 15 had 2 subsequent doses and 13 required 3 or more additional doses. The response lasted 4-12 months in 35 men and in 15 men the response lasted a year or longer.

  • The data from this study are limited to an abstract presentation.


  • Retrospective review.

  • MPA 400mg or 150mg by IM injection in 48 men treated with LHRH agonists and experiencing hot flushes.

  • Before MPA was used, the average number of hot flushes experienced was 4.73, and the severity was 2.19 (1=mild, 3=severe).

  • After MPA treatment, the number had fallen to 1.50 and the severity was reduced to 0.075 (p<0.05 for both changes). No difference between the two doses was seen.

  • The majority of patients (81%) had significant improvements in the severity of their hot flushes and a >50% reduction in the number and severity of hot flushes occurred in 76% and 78% of patients respectively.

  • A median of 4 injections was given over a 43 month period, averaging 1 injection every 10 months.

  • The only IM injection in the UK is the 150mg Depo-Provera, which is given every 12 weeks for contraception.


  • Cross-over study.

  • Megestrol 20mg bd or placebo for 4 weeks, then vice versa, followed by an open-label 4-week treatment period of megestrol 10-80mg/day.

  • Women (n=97) and men (n=66) experiencing hot flushes following treatment for breast cancer or ADT (medical or surgical) for prostate cancer, respectively.

  • The median number of hot flushes per day at baseline was 8.4 (range 2 to 29.1) for the men, and the median hot flush score was 17 for the men.

  • After 4 weeks of megestrol, the median number of hot flushes experienced by the men was 20% of the baseline average vs. 81% of baseline average in the placebo group, p<0.001. The median hot flush scores were 13% vs. 84% of the baseline averages, respectively, p<0.001.

  • A 50% reduction in hot flushes was reported by 79% of men treated with megestrol (26/33) vs. 12% treated with placebo (4/33), p<0.001.

  • No clear answer to which treatment the patients preferred was seen. In the patients who received placebo first, 87% said megestrol was better, but only 45% of those who received megestrol first preferred it, probably because of the delayed effect after the start and the carryover effect after stopping.

  • Cross-over analysis had to be ignored because after stopping megestrol, it took 2-3 weeks for its effects to cease, giving a carry-over effect.

  • This study does not address any potential toxicity issues, e.g. thromboembolism, which may be a risk with higher doses, and the effect of low doses megestrol on the course of hormonally sensitive cancers.


  • Long-term use (three years or longer) following the above short-term study.

  • Megestrol doses <20mg/day to 160mg/day.

  • 59 patients (40 men).

  • The most common dose was 20mg/day.

  • Breakthrough hot flushes were experienced by 17 men and 8 men reported infrequent and mild hot flushes.

  • Episodes of chills were described by 22 of the men, even described as ‘cold flashes’, but these were not as bothersome as the hot flushes.

  • Note that only a megestrol 160mg tablet is available in the UK.


  • Pilot study.

  • Paroxetine 12.5mg/day during week 1, 25mg/day during week 2, 37.5mg/day during week 3 and the dose was either maintained at 37.5mg/day or reduced to 12.5-25mg/day during week 4.

  • 24 men with at least 14 bothersome hot flushes/week following ADT.

  • At baseline, 14 patients reported 4-9 hot flushes a day, 3 reported 2-3 per day and 5 reported ≥10 per day.

  • 18 patients completed the study.

  • The median daily hot flush frequency and hot flush score during the fourth treatment week decreased from baseline by 50% (95% CI 34-92%) and 59% (95% CI 31-87%) respectively.

  • Median hot flush frequencies fell from 6.2 to 2.5/day.

  • Improvements were seen in various QoL measurements, including abnormal sweating, depression, sleeping, anger and hot flush control.

  • The sample size would give 80% power to detect at least a 50% reduction in hot flush scores.

  • Determining the optimal paroxetine dose is difficult: impressive reductions were seen in the first 3 weeks as the dose was increased but longer term data are not available.


  • Pilot study.

  • Paroxetine 10mg/day for 4 weeks.

  • 10 men having at least 7 hot flushes/week following ADT.

  • The mean number of hot flushes experienced at baseline were 3.5 ± 2.6/day and were rated mainly as ‘quite a bit’ or ‘extremely’ severe (mean 4.6 ± 3, 1 = not at all, 5 = intermediate, 10 = extremely severe).

  • After 4 weeks of paroxetine treatment, the frequency of hot flushes was reduced to a mean of 2.0 ± 2.7/day (p=0.009) and the average rating for severity was also reduced to 2 ± 2.7 (p=0.0332).

  • Quality of life was improved, with an increase from the baseline score of 3.8 ± 2.3 to 6.9 ± 2.6 (p=0.0218, 1=worst possible life, 10=best possible life).

  • A lower paroxetine dose than used in the trial above was effective, but longer term data are lacking.


  • Randomised placebo-controlled 8-week study

  • Paroxetine 10mg/day (n=45) or placebo (n=45)

  • Androgen-deprived prostate cancer patients.

  • Significantly lower Hot Flush Index (HFI= frequency x severity) reported after 4 and 8 weeks of paroxetine treatment [week 4: 3.49±0.77 vs. 4.08±1.08 (placebo), p=0.042; week 8: 3.33±0.73 vs. 4.03±1.07, p=0.014 (placebo)]

  • Abstract only available


  • Case reports

  • Sertraline 25mg – 100mg daily

  • 5 men with prostate cancer and experiencing hot flushes.

  • Case 1: Patient was initiated on sertraline 25mg daily which was raised to 100mg daily over 6 weeks. A significant decrease in the frequency (from 5-10 per day to 3-5 per day) and intensity of hot flushes occurred.

  • Case 2: Sertraline 100mg daily improved the quality of life of the patient who had been experiencing a few hot flushes each day and was being woken once or twice a night by the flushes.

  • Case 3: Sertraline 100mg daily reduced the number of hot flushes in a patient who was woken 1-2 times per night because of them. A 50mg daily dose maintained the positive effects.

  • Case 4: Patient experienced hot flushes with drenching sweats several times a day. With a dose of sertraline 100mg daily the frequency of the flushes fell from 8 to 2 per day, with the mean duration reduced from 2-3 minutes to less than 1.

  • Case 5: A patient who experienced distressing hot flushes several times a day and was awakened several times night was treated with sertraline 125mg daily. Although no decrease in the frequency and intensity of the flushes was reported, he did not complain about them as much.

  • These limited case reports show that sertraline is effective in reducing hot flushes, but the data are limited by the lack of detail and small patient numbers; a larger study would be required.


  • Pilot study.

  • Venlafaxine 12.5mg bd for 4 weeks.

  • Five men receiving ADT for prostate cancer and 23 women with a history of breast cancer.

  • The number of hot flushes fell from a baseline average of 6.6 hot flushes per day (range 3.1 to 33.9) to 5.7 after week 1 and to 4.3 (range 0 to 19) at the end of the study period.

  • Fifteen patients reported ≥50% reduction in the number of hot flushes, and the incidence of severe/very severe was reduced from an average of 1.4 to 0.1 per day (p<0.0002).

  • Overall, 19 patients (68%) thought that venlafaxine reduced hot flush activity and 18 (64%) would want to continue with the therapy.

  • Although the patient numbers are too small to draw firm conclusions on subsets, there did not appear to be any differences between the responses of the men and women.


  • Pilot study.

  • Venlafaxine 12.5mg bd for 4 weeks.

  • 21 men receiving ADT and with at least 14 hot flushes/week. 16 men had evaluable data.

  • During the baseline period an average of 10 hot flushes per day were reported (range 0.7 to 33.9), which fell to 7 by week 1 and to 6 (range 0 to 15) by the end of the treatment period.

  • Hot flush activity decreased significantly in all but 5 patients during the first treatment week and by the end of week 4, 13 men reported a reduction in hot flush activity.

  • A number of side effects occurred, including appetite loss, sleepiness, constipation, dizziness, tiredness, dry mouth, abnormal sweating, but the incidence was lower at week 4 than at baseline and three patients discontinued because of adverse gastric reactions, including severe nausea.

  • At the end of the study, 11 of the 21 men who started the study perceived a benefit from venlafaxine and wished to continue taking it.

  • Both of the venlafaxine studies are too short to draw firm conclusions regarding the long-term efficacy of it for hot flushes.


  • Randomised, double-blind, double-dummy 8 week study.

  • Venlafaxine SR 75mg/day (n=102), MPA 20mg/day (n=108) or cyproterone acetate 100mg/day (n=101).

  • 309 men treated with leuprorelin 11.25mg who, after 6 months of treatment, experienced ≥14 hot flushes in the week before enrolment. Patients receiving SSRIs, steroid hormones, clonidine and gabapentin were excluded.

  • The primary endpoint was the change in hot flush score after 4 weeks of treatment. Daily hot flushes were assessed using the validated Mayo Clinic hot flush diary, with the score = number of hot flushes x average severity (mild =1, very severe = 4).

  • At baseline hot flushes were described as: mild (20.1%), moderate (43.4%) and severe (36.6%). The median (mean) weekly number of hot flushes was 33 (44.4). The daily hot flush score was 7.6 (10.5).

  • Results at week 4 are below. Pairwise comparisons of treatment groups confirmed that the decreases in hot flush scores were significantly larger in the cyproterone and MPA groups than in the venlafaxine group. There was no significant difference between the cyproterone and MPA groups.

  • Overall, after 4 weeks of treatment, 70.9% had an improvement of at least 50% in their hot-flush score and 22.7% reported a complete absence of hot flushes. Improvements for venlafaxine were significantly lower than in the other two groups (p=0.0006).

  • EORTC Quality of life questionnaires were completed but the only significant difference was in the emotional functional scale, which was significantly better with venlafaxine.

  • 60-84.1% of patients treated with cyproterone or MPA rated their treatment as good, compared to 28.4 to 33.7% in the venlafaxine group (p<0.0001).

  • The rate of adverse events and discontinuation due to AEs did not differ significantly between the three groups.






Median score at randomisation





4 week median (mean) weekly number of hot flushes

9.0 (15.7)

21.0 (26.6)

4.0 (1.8)

P<0.0001 vs. venlafaxine

7.5 (12.7).

P<0.0001 vs. venlafaxine

4 week median (mean) daily hot flush score

1.6 (3.6)

4.0 (6.5)

0.9 (1.8)

1.3 (2.7)

4 week median relative change in daily hot flush score from baseline



(-74.3 to -2.5)


(-100 to -74.5)


(-98.9 to -64.3)

8 week median score




Median relative change in daily hot flush score from 4 to 8 weeks


(-50 to -38.3)


(-88.9 to 0)


(-96.8 to 0)

Median relative change in daily hot flush score from baseline to 8 weeks


(-80.9 to -21.7)


(-100 to -83.5)


(-100 to -77.8)

Adverse events




AEs related to study med

26 (39.4%)

36 (48%)

18 (30%)

Patients with ≥1 AE leading to discontinuation

7 (6.9%)

8 (8%)

9 (8.4%)

  • All patients received 2 tablets in the morning and one in the evening from weeks 1-8, then one tablet twice a day, to account for venlafaxine dosing and discontinuation (37.5mg/day during weeks 9 and 10).

  • The primary goal of 351 patients, to take into account a 14% dropout rate, was not reached.

  • No placebo group was included due to the availability of data for the study drugs vs. placebo.

  • There was an imbalance in the median hot flush score at randomisation, with a higher score and more patients requesting hot flush treatment in the MPA group; this may have affected the results.

  • Data are only available for 8 weeks of treatment; long-term data on efficacy and safety are needed.

  • Hot flush score: total number of mild hot flushes + 2 x total number of moderate hot flushes + 3 x total number of severe hot flushes + 4 x total number of very severe hot flushes recorder in a given week and divided by the number of days the symptoms were recorded on.


  • Randomised, double-blind, placebo-controlled 12 week study.

  • Placebo + milk powder (n=30), venlafaxine 75mg/day + milk powder (n=30), placebo pill + soya powder 20g/day (n=30) or venlafaxine 75mg/day + soya powder 20g/day (n=30).

  • Androgen-deprived prostate cancer patients, experiencing ≥4 moderate-severe hot flush/day. Patients receiving concurrent therapy for hot flushes, or recent use of venlafaxine, MAOI, SSRI SNRI were excluded.

  • The primary endpoint was hot flash severity score (HFSS) and the trial was powered to detect a 12.5% difference in HFSSS between placebo and treatment.

  • HFSSS decreased significantly over time in each arm (p<0.001) but by week12 there were no significant differences among the treatment arms. The initial effect of venlafaxine was lost over time; at week 2, hot flash severity score decreased by 28% for venlafaxine vs. 2% for placebo (p<0.005), however, by week 12, the decrease was 29% for venlafaxine vs. 36% for placebo (p=0 .723).

  • The number of vasomotor symptoms decreased significantly in all arms (p <0.001) but there were no significant differences among treatment arms at any time. Although participants in the venlafaxine arm tended to have fewer hot flashes during the initial 2 weeks, this early difference had disappeared by week 12.

  • Hot flash severity decreased significantly in each arm (p<0.001). There were no significant differences in the comparison of soy and placebo at any time. The venlafaxine arm tended to have lower hot flash severity values at weeks 1-4 but the difference was not significant at week 12.

  • Overall, neither soy protein nor venlafaxine were effective in treating hot flashes in men over a 12-week period. The initial benefit of venlafaxine did not continue past week 4.

  • The sample size of 176 patients to maintain trial power was not obtained.

  • The trial was stopped early by the Data Safety Monitoring Board because of lack of effect.


  • There are a number of treatments which have been studied for the treatment of hot flushes in men undergoing androgen deprivation therapy, most are unlicensed for this indication, such as clonidine, gabapentin, medroxyprogesterone acetate, megestrol and SSRIs, while cyproterone is licensed for this.

  • There are few well designed, randomised studies. Some data come from smaller open-label or pilot studies or case reports, which were not placebo-controlled and have not been well reported.

  • Many of the studies are short, do not carry the strength of a longer, prospective, randomised, placebo-controlled trial and may prevent firm conclusions being drawn.

  • Optimum doses of the individual drugs may be difficult to determine, and it may be a case of starting with a low dose and titrating upwards until hot flushes are controlled, or until adverse events lead to discontinuation. Bear in mind that patient variability means that one specific treatment may not be suitable for every patient.


  • Of the treatments listed in the Q&A, only cyproterone is licensed for the treatment of hot flushes in men who are treated with LHRH agonists or who have undergoing orchiectomy.(13)

  • Interactions, contraindications, precautions, adverse events etc. of each therapy are not discussed in this Q&A, but information can be found via the electronic Medicines Compendium.

Quality Assurance
Prepared by

Varinder Rai, Medicines Information Manager, London Medicines Information Service (Northwick Park Hospital) (Based on earlier work by Alexandra Denby)

Date Prepared

August 2015

Checked by

Lekha Shah, Principal Pharmacist, London Medicines Information Service (Northwick Park Hospital)

Date of check

September 2015

Email contact
Search strategy

  • Embase: *PROSTATE CANCER/ AND HOT FLUSH/[Limit to: Human and English Language and Publication Year 2013-2015]

  • Medline: HOT FLASHES/ AND PROSTATIC NEOPLASMS/ [Limit to: Publication Year 2013 – Current]


Available through NICE Evidence Search at

Share with your friends:

The database is protected by copyright © 2019
send message

    Main page