The use of chemotherapy gave a small, but statistically significant, increase in the time to progression of the disease (p = 0.02²⁴⁵; p = 0.018²⁴⁶).

None of the three mitoxantrone studies showed a significant increase in overall survival^245-247.

The major toxicities associated with mitoxantrone included grade 3-4 neutropenia (7%), nausea and vomiting, alopecia (24%) and cardiotoxicity (66%)²⁴⁷. In another study, grade 3-4 cardiotoxicity appeared in 5% of patients who received mitoxantrone, and haematological toxicity was significantly higher in patients who received the chemotherapy²⁴⁵.

The docetaxel efficacy results were compared with mitoxantrone (the chemotherapy reference at the time) in the 2004 Tannock et al study²⁴⁴, which included 1,006 men with AIPC. There were two different docetaxel administration regimes: some patients received a dose of 75 mg/m2 every 3 weeks and others a weekly dose (30 mg/m2/week over 6 weeks). The mitoxantrone was administered at doses of 12 mg/m2 every 3 weeks.

When docetaxel was compared with mitoxantrone, the mortality hazard ratios were: (HR = 0.76 [95% CI: 0.62-0.94]; p = 0.009) for docetaxel every 3 weeks and (HR = 0.91 [95% CI: 0.75-1.11]; p = 0.36) for the weekly scheme. This showed a significant improvement in overall survival with the 3 week docetaxel regime, compared with the mitoxantrone (24% reduction in the risk of death).

A significant reduction was also noted in pain in patients who received the 3-week docetaxel regime compared with mitoxantrone (35% vs 22%; p = 0.01) but not with the weekly programme (31 %). The median pain response duration (3.5 vs 5.6 months) was not significant different between the groups.

The quality of life also showed a significant improvement in patients treated with the 3-week docetaxel regime compared with those treated with mitoxantrone (22% vs 13%; p = 0.009).

Degree 3 and 4 neutropenia in patients included in the 3-week regime was statistically significantly more frequent than those who received weekly docetaxel or mitoxantrone (32% vs 2% vs 22%), although the frequency of febrile neutropenia was less than 4% in all groups.

There was a high incidence of nausea and vomiting in all programmes (38% to 42%). Diarrhoea was significantly more frequent in the docetaxel regimes.

Interruption of treatment with docetaxel was due to fatigue, musculoskeletal events, changes in the nails, sensory neuropathy and infection. In the mitoxantrone group, the main reason was due to cardiac dysfunction.

The combination of docetaxel and oestramustine has also been compared with mitoxantrone in patients with AIPC. The Petrylak et al study²⁴³ administered docetaxel to one group (60 mg/m2 on day 2) then oestramustine (280 mg/m2 days 1-5) and mitoxantrone (12 mg/m2 on day 1) to another group. In the Oudard et al study²⁴⁸, there were 3 branches of treatment: one that received docetaxel at a dose of 70 mg/m2 (administered on day 2 every 3 weeks) and oestramustine (280 mg/m2 administered 3 times a day on days 1-5); another that received docetaxel at a dose of 35 mg/m2 (days 2 and 9, repeated over a 3 week cycle) and oestramustine (as above), and another that received mitoxantrone (12 mg/m2 every 3 weeks).

In the Petrylak et al study there was a significant improvement in overall survival for those treated with docetaxel and oestramustine (17.5 months vs 15.6 months; p = 0.02)²⁴³. However, in the Oudard et al study, although the median overall survival was greater for those treated with docetaxel (18.6 and 18.4 months) than for mitoxantrone (13.4 months), there were no significant differences between the regimes (p = 0.3)²⁴⁸.

Regarding disease progression, significant improvements were found for those receiving combination treatment [6.3 months v 3.2; p <0.001 in one study²⁴³; p <0.00001 in the other²⁴⁸].

The percentage of patients who achieved a PSA response (at least a 50% reduction in levels) was statistically significantly better in patients treated with docetaxel and oestramustine [50% vs 27%; p <0.001 in one study²⁴³; p <0.00001 in the other²⁴⁸].

With regard to pain relief, the Petrylak et al study found no significant differences between the two groups when evaluated by patients²⁴³. However, the Oudard et al showed a statistically significant improvement in pain index for each of the two docetaxel groups (70 mg/m2 and 35 mg/m2) when compared with mitoxantrone (40% and 29% vs 17%)²⁴⁸.

There was a significant improvement in the ECOG performance status of patients treated with docetaxel compared with mitoxantrone (60% and 48% vs 28%, respectively).²⁴⁸

With the docetaxel-oestramustine combination, there were more gastrointestinal side effects (p = 0.001), nausea and vomiting (p = 0.001), infection (p = 0.004), metabolic toxicity (p <0.001) and neurological dysfunction (p = 0.001 )²⁴³. In addition, there was oestramustine-induced thrombosis in 7% of the patients treated with docetaxel, despite receiving anticoagulant treatment²⁴⁸.

Granulocytopenia grade 3 and 4 was the most common toxicity in patients treated with docetaxel 70 mg/m2 and mitoxantrone (37% and 48% respectively), but this was not seen with the lower dosage of docetaxel²⁴⁸.

Comparison of docetaxel-oestramustine v docetaxel in patients with androgen-independent prostate cancer (AIPC) and metastatic prostate cancer was evaluated by the Eynard et al study²⁴⁹, which applied docetaxel (70 mg/m2 on day 2) and oestramustine (280 mg/m2 administered twice daily on days 1-5) in a group of patients (n = 47) and docetaxel (75 mg/m2 on day 1) in the other (n = 44).

There was a statistically significant difference in the PSA response (decrease in PSA level in ≥ 50% maintained for ≥ 3 weeks), which was: 68% [95% CI: 55-81] v 30% [95% CI: 16-43]. The median PSA response duration was 6.0 months in both groups.

The median time to progression was 5.7 months [95% CI: 4.7-6.8] vs 2.9 months [95% CI: 2.0-6.9], and the median survival time was 19.3 months [95% CI: 14.6-25.9] vs 17.8 months [95% CI: 11.8-20.9], both without significant differences.

Haematological and non-haematological toxicity plus the quality of life were similar in both groups.

6% of patients treated with the combination had phlebitis, possibly due to the oestramustine. One patient in each group decided to discontinue the study because of toxicity. One treatment-related death (pulmonary oedema) occurred before 30 days in the docetaxel group²⁴⁹.

The European Association of Urology clinical practice guideline⁴ recommends establishing a chemotherapy regime for AIPC patients who have two consecutive PSA increases above the reference values, and a PSA level above 5 ng/ml. It also recommends the decision for the start of the chemotherapy treatment should be done on an individual basis.

There is consistent evidence from randomised trials that treatment with bisphosphonates increases bone density in the spinal column in men receiving hormone therapy for prostate cancer. In patients treated with bisphosphonates, an average increase in bone density of 1-5% was seen in the first year of hormone treatment. However, a significant reduction of 0.4 - 4.9% was seen in those who received placebo or the standard treatment during the same period. The bisphosphonate group was about 5% greater^254-258.

In patients with androgen-independent prostate cancer (AIPC), when compared with placebo, bisphosphonates achieved a modest reduction in skeletal events (such as the occurrence of pathological fractures, spinal compression, or the need for surgery or radiotherapy treatment for bone metastases): 37.8% vs 43.0%; absolute risk reduction of 5.2%²⁵².

Saad et al studied the use of zoledronic acid in patients with AIPC, with a high loss rate (ranging between 62% and 72% depending on the treatment group). Zoledronate at a 4 mg dose caused a statistically significant decrease in the proportion of patients with skeletal events when compared with placebo. However, the difference with zoledronic acid at a dose of 8 mg (subsequently reduced to 4 mg) vs placebo was not significant²⁵⁹. In addition, zoledronate reduced the incidence of skeletal events by 36% (RR = 0.640; p = 0.002)²⁶⁰. This decrease was highest in patients without pain261. The bisphosphonate delayed the first skeletal episode by more than 5 months (p = 0.009, which was a significant difference when compared with placebo)²⁶⁰. A significant RR was seen for the proportion of patients with a skeletal event (RR = 0.71 [95% CI: 0.50-0.99]) when zoledronic acid was compared with placebo²⁶².

When comparing zoledronic acid with placebo or the standard treatment in patients with AIPC, symptomatic fractures did not appear in the year following the start of hormone therapy. As for asymptomatic fractures, there were no differences in the rates for both groups²⁵⁸. In another study with the same design, the relative risk for the proportion of patients with pathological fractures was significant: RR = 0.57; [95% CI: 0.38-0.88]²⁶².

In patients with AIPC, the rates of spinal compression, bone surgery and bone radiotherapy did not differ significantly when comparing bisphosphonate and placebo²⁵².

In men with AIPC, there was a non-significant trend towards better results with bisphosphonates when compared with placebo for pain relief in bone metastasis²⁵².

The use of zoledronic acid at a dose of 8 mg in AIPC produces an improvement in the average pain rating at 15 months of treatment when compared with placebo (p = 0.026), but there were no significant differences when comparing bisphosphonate at a dose of 4 mg to placebo (p = 0.134). There were no significant differences in analgesia levels when comparing each of the bisphosphonate treatments with placebo²⁵⁹.

In men with AIPC, zoledronic acid produced significant reductions in bone pain in the long term, when compared with placebo²⁶⁰.

The use of bisphosphonates in patients with androgen-independent prostate cancer resulted in a decrease in consumption of painkillers when compared with placebo²⁵².

In patients with AIPC, the median survival time was 464 days for patients treated with placebo, 546 days for patients who received zoledronate at 4 mg (p = 0.091), and 407 days for patients who received a dose of 8 mg (p = 0.386)²⁵⁹.

In men with metastatic prostate cancer treated with androgen deprivation, no significant differences were found in the rate of severe adverse effects when comparing bisphosphonates with placebo^255,257,258.

In patients with AIPC, zoledronic acid resulted in a deterioration of kidney function: 15.2% of patients treated at a dose of 4 mg and 20.7% of those who received a dose of 8 mg, with 11.5% of those treated with placebo²⁵⁹.

In patients with AIPC, the quality of life did not differ significantly when comparing bisphosphonates and placebo²⁵².

A systematic review in 2007²⁶³ found only 26 cases of mandibular osteonecrosis in patients treated with bisphosphonates, which had been previously reported. Of the 26 cases found, 87% occurred in women, 78% older than 60 years. For 80%, in the area of osteonecrosis, dental damage already existed or the patients had received treatment prior to surgery. There was no clear link found between the duration of treatment with bisphosphonates and the appearance of mandibular osteonecrosis. It should be remembered that the frequency of this adverse effect is very low: only 1 case appeared in a series of more than 7,000 women treated for 3 years with zoledronic acid. The estimated incidence is 1 case per 10,000-100,000 inhabitants/year in patients treated with bisphosphonates.

When comparing the reduction of pain reported in patients treated with Sr-89 v placebo, no significant differences between the two treatments were found in the long term (1-3 years), but there were differences in the short term (5 weeks)^266,267. When Sr-89 was compared with local external beam radiotherapy (ERT), some studies found less pain in the group treated with Sr-89 + radiotherapy (RT), although in others there were no differences^268,269. When Sr-89 + local RT was compared with local RT, the reported pain was similar in both groups, but the appearance of new painful locations was significantly higher in the group receiving external radiotherapy^270,271.

It seems that the Sr-89 is effective for pain control in bone metastasis in up to 70% of patients¹⁷.

The Sartor et al study²⁷² noted that the use of Sm-153 has positive effects on pain for 1-4 weeks after starting treatment, when compared with placebo (correlation coefficient r = 0.78; p <0.0001). In addition, it decreased the use of opioids 3-4 weeks after starting the treatment (p <0.0284).

When compared with placebo, Sm-153 achieved pain reduction in a greater proportion of patients after starting the treatment (38% vs 18%; p = 0.008). The same occurred 4 weeks after starting treatment (55% vs 35%)²⁷².

When comparing Sr-89 with local ERT, biochemical progression-free survival was comparable between the two groups, while overall survival was significantly greater in the group receiving ERT²⁷⁰. However, when the same comparison was performed in a different trial, overall survival was similar in both groups²⁷¹.

When comparing Sr-89 with placebo, the group treated with Sr-89 had a better overall survival at 2 years²⁶⁷. When Sr-89 + local RT was compared with local RT, no differences in global survival were found²⁶⁸. However, when Sr-89 + chemotherapy (CT) was compared with CT, better results for overall survival were found in the Sr-89 group²⁷³.

Sr-89 was associated with haematological toxicity (thrombocytopenia, neutropenia) in approximately 30-50% of patients who received it (usually moderate, grade ≤ 2)¹⁷.

In randomised clinical trials that compared Sr-89 v local RT, the rate of adverse effects (haematological toxicity, nausea and vomiting) was similar in both groups^270,271.

The only statistically significant side effect associated with Sm-153 in a trial was temporary and slight myelosuppression²⁷². The European Association of Urology clinical practice guideline found that early use of radioisotopes may make it harder for the administration of chemotherapy, because they cause myelosuppression⁴.

When comparing Sr-89 + local RT v local RT, no significant differences were found regarding the quality of life²⁶⁹.