Iloprost: new indication. Not adequately assessed

(1) Iloprost, a vasodilatory prostacyclin analogue administered by infusion, is indicated for second-line therapy in patients with severe ischaemia of the lower limbs, when surgical revascularisation fails or is contraindicated. (2) In thromboangiitis obliterans the clinical file on iloprost has remained inadequate since the product was first released. (3) A meta-analysis of 6 clinical trials giving conflicting results in patients with stage III or IV lower-limb arterial disease favoured iloprost. But the results of this meta-analysis are uninterpretable because of methodological biases. It is not known what effects iloprost has in the short term (on pain and skin damage) or in the long term (on the risk of amputation). (4) The iloprost dose must be adjusted individually according to adverse effects linked to vasodilation.     

Alteplase: new indication. Inadequately assessed in ischaemic stroke

(1) Alteplase is the first thrombolytic drug to be approved in France for the treatment of ischaemic stroke within three hours of symptom onset. (2) The clinical evaluation dossier contains nine placebo-controlled trials, of which six were relatively large. In the two NINDS trials (624 patients in total), treatment was started within the first three hours and it showed no survival benefit. Near-complete functional recovery was more frequent in the alteplase group than in the placebo group. In the two ECASS trials (620 and 800 patients) and the ATLANTIS trials (142 patients and 613 patients), treatment was started within the first six hours and it showed no significant benefit in terms of survival or functional recovery. (3) There are two meta-analyses of these trials. They confirm the lack of a survival benefit with alteplase. Using a combined endpoint, one meta-analysis showed that treating 1000 patients with alteplase prevented death or major disability (dependency) in 55 patients. The other meta-analysis underlined the importance of a short interval between the onset of symptoms and the beginning of treatment. (4) Intracranial haemorrhage is the most important adverse effect. One meta-analysis showed that alteplase caused 62 additional symptomatic intracranial haemorrhages (including 25 deaths) per 1000 treated patients. (5) Various retrospective subgroup analyses have tentatively identified subgroups of patients at a particularly high risk of adverse effects, but subgroup analyses provide only weak evidence. The patients most likely to benefit from alteplase, started within three hours of symptom onset, remain to be defined. (6) The current health infrastructure in France would allow only a small number of stroke patients to be treated with alteplase under the kind of conditions prevailing in clinical trials (imaging to confirm ischaemic stroke, and treatment very soon after the onset of symptoms). (7) In practice, there is a narrow margin between the wanted and unwanted effects of alteplase. This treatment should be used only by specialised teams and for strictly selected patients. Research must continue, particularly to identify those patients most likely to benefit from alteplase, and those most likely to be harmed in whom thrombolysis is contraindicated.     

Trastuzumab: new indication. Adjuvant breast cancer treatment for a minority

An improvement in survival time, but cardiotoxicity remains a concern.     

Paroxetine: new indication. In generalised anxiety: uncertain efficacy

(1) Generalised anxiety is defined as overwhelming anxiety lasting at least 6 months. (2) Psychological treatment should be tried first. When the patient fails to cope the first-line drug is a benzodiazepine such as diazepam, prescribed for a short period. (3) In France, generalised anxiety has now been added to the licensed indications of paroxetine, a "selective" serotonin reuptake inhibitor antidepressant. Two placebo-controlled trials lasting 8 weeks showed a moderate improvement on the Hamilton anxiety score. Another trial showed no significant difference between the paroxetine and placebo groups. The clinical relevance of this improvement is unclear, however, and the trials suffered from methodological biases. Paroxetine has not been reliably compared with benzodiazepines, psychotherapy or buspirone in patients with generalised anxiety. (4) One trial showed 11% relapse in the paroxetine group and 40% in the placebo group during the 6 months following paroxetine withdrawal, among patients who had initially responded; once again, however, methodological flaws undermine these data. (5) The adverse effect profile of paroxetine in generalised anxiety is similar to its profile in other patients: in particular potentially serious drug interactions and withdrawal symptoms when treatment is stopped abruptly. (6) In practice, the standard drug therapy for generalised anxiety is a benzodiazepine such as diazepam. Paroxetine, whose clinical efficacy remains to be established in this setting, offers no tangible therapeutic advance.     

Paroxetine: new indication. In social phobia: minimal assessment

Social phobia is generally defined as an intense and persistent fear of one or several social situations, with important repercussions for occupational activity or social life. Cognitive psychotherapy and antidepressants have partial efficacy. There is no reference drug therapy. In France, paroxetine is the first drug to be granted a licence for patients with social phobia. Clinical evaluation consists of data from four placebo-controlled trials lasting only 12 to 24 weeks. Treatment with paroxetine was associated with a significant improvement in standard social-phobia scores, although most patients remained symptomatic. Paroxetine is the best assessed selective serotonin reuptake inhibitor in this setting. However, long-term data are lacking, and the disorder is chronic. Paroxetine has not been compared with cognitive therapy. About one-third of patients in clinical trials stopped taking paroxetine, mainly because of adverse events. Gastrointestinal upset, sleep disturbance and ejaculatory problems are frequent. Paroxetine also has the potential to interact with other drugs. In practice, paroxetine may help some patients, provided they are aware of its limitations. The long-term effects of paroxetine in this setting remain unknown.     

Anastrozole: new indication. Adjuvant treatment of non metastatic breast cancer: useful for some patients

(1) For postmenopausal women with hormone-receptor-positive breast cancer, the reference adjuvant treatment after surgical excision is tamoxifen (an anti-estrogen), taken orally at a dose of 20 mg/day for 5 years. (2) Anastrozole is the first aromatase inhibitor to be licensed for this use in France. (3) Marketing authorisation was based on the short-term results of a double-blind trial comparing anastrozole (1 mg/day) with tamoxifen (20 mg/day) in 9366 women. The trial is planned to last five years. The results obtained after median follow-up of 4 years showed no difference between the groups in overall survival (109 deaths in each group). But first pathological events were significantly less frequent in the group taking anastrozole (13% versus 15%). Note that these results are undermined by a number of methodological flaws, including relatively short follow-up and definition of relapses using an endpoint mixing heterogeneous prognostic factors. (4) Musculoskeletal disorders, fractures (7.1% versus 4.4%) and hypercholesterolemia were statistically more common with anastrozole than with tamoxifen. Women taking anastrozole found their sex lives less satisfactory than women taking tamoxifen. The following adverse events were statistically less common with anastrozole than with tamoxifen: hot flushes (35.0% versus 40.3%), metrorrhagia, venous thromboembolism (1.1% versus 1.8%), ischaemic stroke (1.1% versus 2.3%), and endometrial cancer (3 versus 15 cases at 4 years). (5) In practice, anastrozole may be beneficial for women who cannot use tamoxifen, such as those at high risk of thrombosis. Anastrozole costs ten times more per day than tamoxifen. Tamoxifen remains the reference adjuvant treatment for all other women.     

Gabapentin: new indication. In postherpetic neuralgia when amitriptyline fails

(1) Postherpetic pain is infrequent, but the incidence increases with age. (2) The reference treatment for postherpetic pain is oral amitriptyline or desipramine. (3) Gabapentin, an antiepileptic agent, is the first drug to be granted specific approval in France for the treatment of postherpetic pain. (4) In two placebo-controlled trials, gabapentin at a dose of between 1 800 and 3 600 mg/day halved the intensity of pain in about one in three patients. In comparison, pain improved in about 50% of patients taking amitriptyline in clinical trials. (5) Both gabapentin and amitriptyline provoke sedation, but dizziness and peripheral oedema are more frequent on gabapentin, while atropinic effects predominate with amitriptyline. (6) Daily treatment is 10 times more costly in France. (7) In practice, the standard treatment of postherpetic pain remains oral amitriptyline or desipramine. Gabapentin is an alternative, given its different safety profile.     

Celecoxib: new indication. Colorectal cancer: no preventive benefit

(1) Familial adenomatous polyposis is a genetic disorder associated with multiple adenomatous colorectal polyps that invariably progress to colorectal cancer. Gastroduodenal polyposis and extra-gastrointestinal desmoid tumours are other major sources of morbidity in these patients. (2) The current strategy used to prevent colorectal cancer in patients with APC gene mutations consists of yearly monitoring starting in adolescence, and prophylactic colectomy in early adulthood if polyposis occurs. (3) On the basis of pathophysiological, experimental and epidemiological evidence, some specialists have postulated that certain nonsteroidal antiinflammatory drugs (NSAIDs) might have a preventive effect on colorectal adenomas and cancer. (4) Aspirin and sulindac were tested for the prevention of polyps in patients with familial adenomatous polyposis, with uncertain results and weak evidence of effectiveness. (5) Celecoxib was tested in a comparative randomised double-blind trial lasting 6 months. It involved 77 patients with familial polyposis and colorectal polyps, and 6 patients with only duodenal polyps. On the basis of composite endoscopic criteria, a celecoxib dose of 800 mg/day (but not 200 mg/day) reduced the number and surface area of adenomatous colorectal polyps in patients with familial adenomatous polyposis. It is not known whether celecoxib also reduced the risk of colorectal cancer. A global qualitative analysis suggested that celecoxib was also effective in reducing duodenal polyps. (6) Nearly one-third of patients receiving celecoxib 800 mg/day in this trial developed rectal bleeding. Another preventive trial was stopped when an excess of cardiovascular events was found in patients taking celecoxib. (7) The long-term risk-benefit balance of celecoxib 800 mg/day is not known nor whether efficacy persists after treatment discontinuation. (8) In practice, it is better not to use celecoxib to prevent colorectal cancer: its efficacy has not been demonstrated, even in familial polyposis, and it carries a major risk of bleeding and cardiovascular events.     

Clopidogrel: new indication. In combination with aspirin: marginal additional benefits

In patients with myocardial infarction who are not eligible for angioplasty, adding clopidogrel to aspirin reduces the overall 15-day mortality rate, but the subsequent outcome is not known.     

Doxetaxel: new indication. Prostate cancer: a few more weeks

(1) The standard treatment for metastatic prostate cancer is hormone therapy, based on medical castration (with an LH-RH agonist) or surgical castration (pulpectomy), possibly combined with an androgen antagonist. For patients with hormone-resistant disease the only cytotoxic agents approved in France, estramustine and mitoxantrone, have no proven impact on survival. (2) Docetaxel is now approved in Europe for the treatment of hormone-resistant metastatic prostate cancer, in combination with a steroid. (3) In an open-label comparative trial involving 1006 patients, docetaxel infusion at a dose of 75 mg/m2 every 3 weeks, in combination with prednisone (or prednisolone), significantly extended the median survival time by about 2.5 months as compared with a mitoxantrone-prednisone combination (18.9 versus 16.5 months). In another open-label comparative trial involving 674 patients, a combination of docetaxel + estramustine was significantly more effective than a mitoxantrone + prednisone combination in extending median survival time (17.5 versus 15.6 months). (4) The adverse effects of docetaxel + prednisone were the same as those seen with other indications (hair loss, nausea and vomiting, diarrhea, neutropenia, nail disorders, neuropathies), and were severe in 25% of patients. (5) In France the cost of docetaxel therapy for hormone-resistant metastatic prostate cancer is more than 1000 euros every three weeks. (6) In practice, docetaxel is the first cytotoxic agent shown to prolong survival in men with hormone-resistant metastatic prostate cancer. The benefit is limited, however, especially given the potentially severe adverse effects of docetaxel, which must be disclosed to patients.     

Verteporfin: new indication. In certain age-related macular degeneration: too little evidence

(1) The standard treatment for neovascular forms of age-related macular degeneration is thermal coagulation by argon or krypton laser, but it is only suitable for patients with juxtafoveal or extrafoveal neovascularisation. (2) Verteporfin, a photosensitising agent, is injected intravenously 15 minutes before non thermal red laser activation. This protocol was initially restricted to patients with predominantly visible neovascularisation in the subfoveal choroidal region. In Europe (but not in the United States), this treatment has now been approved for patients with occult subfoveal neovascularisation, i.e. poorly defined zones of fluorescein diffusion. (3) Use in this indication is supported by a single double-blind trial comparing red laser + verteporfin with red laser + placebo. After one year's treatment there was no difference between the groups in the number of patients whose loss of visual acuity had stabilised. One of the several statistical analyses showed that red laser + verteporfin was significantly more effective than red laser + placebo at two years, after five treatment sessions on average. Only a minority of patients benefited, however, and these cannot yet be identified before treatment. It is unclear whether efficacy persists in the long term. (4) Some patients experienced a transient visual deterioration. Verteporfin infusion can provoke local reactions and thoracic or lumbar pain. Cutaneous photosensitisation can occur for 48 hours after treatment. (5) Treatment with verteporfin is expensive. (6) In practice, there is currently no reason to use red laser + verteporfin in patients with age-related macular degeneration and occult neovascularisation, other than in clinical trials.     

Trastuzumab: new indication. Metastatic breast cancer, in combination with docetaxel: promising, but more evaluation is needed

(1) There is no consensus on the optimal chemotherapy for metastatic breast cancer. Patients who have never previously received chemotherapy are generally given an anthracycline-based combination of cytotoxic agents. Options for patients who have already received an anthracycline include a taxane such as paclitaxel or docetaxel. The median survival time with these treatments is only about 2 to 2.5 years. (2) Trastuzumab is a monoclonal antibody directed against HER-2, a protein overexpressed by certain tumours, including about 25% of breast tumours. In 2000, the approved indications included first-line treatment of metastatic breast cancer in combination with paclitaxel. One clinical trial had shown, albeit with a low level of evidence, a median increase in survival of about 4 to 5 months. Trastuzumab is now approved for first-line treatment of metastatic breast cancer, in combination with docetaxel. (3) Evaluation data include the results of an open-label trial comparing docetaxel + trastuzumab with docetaxel monotherapy in 186 patients. The median survival time was significantly longer with the combination (31.2 versus 22.7 months). There are no relevant comparisons with other widely used cytotoxic drugs. Indirect comparison suggests that survival is similar with docetaxel + trastuzumab and paclitaxel + trastuzumab. (4) Data on the trastuzumab-docetaxel combination confirm the known adverse effects of trastuzumab, which include heart failure and diarrhea. Trastuzumab increases the frequency of docetaxel-induced neutropenia, which carries a risk of infections. (5) In summary, the results of clinical trials show that median survival time is increased by a few months when trastuzumab is added to a cytotoxic drug. However, the best cytotoxic agent is not known, and adverse effects are poorly documented. (6) In practice, trastuzumab has only been shown to benefit a minority of women with breast cancer, namely those whose tumours overexpress HER-2. Trastuzumab therapy is an option for metastatic breast cancer treatment, provided patients are enrolled in studies designed to answer the many outstanding questions.     

Rituximab: new indication. In aggressive non Hodgkin lymphoma: benefits must be confirmed

The first-line treatment for diffuse large-B-cell non Hodgkin's lymphoma, a highly malignant lymphoma, is CHOP chemotherapy (cyclophosphamide + doxorubicin + vincristine + prednisone). Rituximab, a monoclonal antibody targeting certain B cells, has received a new indication in the treatment of this type of lymphoma, in combination with the CHOP protocol. In late 2002, the only available evaluation data came from one comparative, unblinded trial in patients over 60 years of age. Addition of rituximab to the CHOP protocol increased both the overall two-year survival rate (70% versus 57%), and the two-year event-free survival rate. Other trials are underway. In this trial, 9% of patients had major systemic reactions during the first rituximab infusion (respiratory disturbances, chills, fever and hypotension). These reactions did not occur during subsequent infusions. About 6% of patients had serious cardiac arrhythmias. In practice, the CHOP protocol remains the standard treatment for aggressive non Hodgkin's lymphoma. Pending further information, addition of rituximab to the CHOP protocol may be justified for patients who meet the inclusion criteria used in the only available clinical trial.     

Citalopram: new indication. No advantage

(1) Citalopram, a serotonin reuptake inhibitor antidepressant, now has a new licensed indication, in the preventive treatment of panic attacks. In France, clomipramine, a tricyclic antidepressant, and paroxetine, another serotonin reuptake inhibitor, are already approved for this use. (2) In the only available comparative trial the efficacy of citalopram (20-60 mg/day) was similar to that of clomipramine (60-90 mg/day). (3) The safety profile of citalopram is different from that of clomipramine. (4) There are no data clearly comparing citalopram with paroxetine in terms of efficacy, safety, drug interactions, or convenience. (5) Clomipramine is much cheaper than citalopram.     

Goserelin and locally advanced prostate cancer: new indication. Pros and cons

(1) Goserelin, a GnRH agonist, has a new licensed indication in France, as an adjuvant to external radiotherapy for locally advanced prosate cancer. (2) The clinical file in this indication includes two trials of satisfactory methodological quality comparing radiotherapy + goserelin with radiotherapy alone. (3) In these trials the radiotherapy + goserelin combination increased the specific-symptom-free survival time. (4) In one trial goserelin caused endocrine disorders in 19% of patients. There were also more cases of urinary incontinence (13% in absolute values) among patients receiving the radiotherapy + goserelin combination. Furthermore, goserelin almost always causes impotence and reduced libido.     

Leflunomide: new indication. In psoriatic rheumatism: too many risks, too little efficacy

(1) The severity of joint involvement in psoriatic rheumatism varies greatly and its outcome is difficult to predict; some patients have long-term spontaneous remissions. The best-evaluated slow-acting treatments are sulfasalazine and methotrexate; adding etanercept can help some patients who do not respond to these drugs. (2) Leflunomide, an immunosuppressant drug, is already marketed for the treatment of rheumatoid arthritis, a condition for which it shows a less favourable risk-benefit balance than methotrexate. (3) Leflunomide is now licensed in France for "active psoriatic rheumatism". (4) The only available clinical data come from a double-blind placebo-controlled trial in 190 patients. On the basis of a combined outcome measure, significantly more patients responded to leflunomide than to placebo (59% versus 29.7%). However, the patients' global assessment was less positive: 15.8% of patients felt their condition had deteriorated during leflunomide therapy, compared to 24.2% of patients in the placebo group. The study population was too heterogeneous to show which types of patients might benefit most from leflunomide therapy. (5) Pharmacovigilance studies have confirmed some severe adverse effects (hepatic, cutaneous and haematological) and have uncovered other previously unrecognised effects such as interstitial pneumonia, hypertension, weight loss, and peripheral neuropathies. (6) In France, leflunomide treatment costs nearly 10 times more than methotrexate. (7) We conclude that leflunomide should not be used to treat psoriatic rheumatism.     

卡培他滨治疗晚期乳腺癌60例临床观察

目的观察卡培他滨（希罗达）对经蒽环类和（或）紫杉类药物治疗后复发或转移性乳腺癌的疗效分析和毒副反应。方法卡培他滨1250mg/m2,2次/d,餐后服用,连续服用2周,休息1周,至少2周期后评价疗效。结果全组60例无CR,PR16例（26.7%）,SD32例（53.3%）,PD12例（20%）,总有效率26.7%,中位无进展生存时间9.8个月。常见副作用为手足综合征,皮肤色素沉着,腹泻,口腔发炎,恶心呕吐,白细胞及血小板减少,大部分为I-II度。结论应用卡培他滨单药治疗蒽环类和（或）紫杉类治疗后复发或转移性乳腺癌疗效确切,毒副反应轻,应用方便,值得临床推广使用。     

Docetaxel: new indication. First-line treatment of non small-cell lung cancer: no advance

(1) The reference first-line drug therapy for patients with non-operable non small-cell lung cancer is a combination of two cytotoxic agents, one of which is a platinum compound. The survival benefit is no more than a few months. (2) The docetaxel + cisplatin combination has now been authorised in France for first-line treatment of locally advanced and metastatic non small-cell lung cancer. Evaluation data includes the results of three comparative trials. (3) In one trial the docetaxel + cisplatin combination was no more effective than the docetaxel + carboplatin combination or the vinorelbine + cisplatin combination on either the survival time (9.4 to 11.3 months) or on other endpoints. (4) Similar results were obtained in a trial versus paclitaxel + cisplatin and gemcitabine + cisplatin (median survival time 8 months in each group). (5) In a trial versus vindesine + cisplatin, the median survival time was longer with docetaxel + cisplatin (11.3 versus 9.6 months). (6) It is difficult to analyse adverse effects in these unblinded trials. Globally, the docetaxel + cisplatin combination did not appear to be safer than the comparator combinations, particularly with regard to serious events. (7) Docetaxel, like paclitaxel, is infused intravenously every three weeks. The comparator combinations tested in the three clinical trials are infused once a week. (8) In practice, for first-line treatment of inoperable non small-cell lung cancer, the docetaxel + cisplatin combination is simply one of several options, and offers no advantages in terms of survival or adverse effects.     

Olanzapine: new indication. New indication in acute mania: just another neuroleptic

(1) Lithium is the first-line treatment for patients with acute mania. For patients with psychosis or intense agitation, an oral neuroleptic can be added (haloperidol or chlorpromazine, the best-assessed drugs of this class). (2) The licensed indications for oral olanzapine, a neuroleptic, explicitly mention the treatment of acute mania. (3) The clinical evaluation dossier on olanzapine in this setting (10 mg to 15 mg/day) is not particularly impressive. In particular, clinical trials included patients with a variety of associated psychotic symptoms. (4) The only comparative trial against another neuroleptic, haloperidol at a high starting dose (10 mg), showed that olanzapine was no more effective. The same applies to a trial comparing olanzapine with disodium valproate. (5) One placebo-controlled trial tested olanzapine as an additional treatment in patients who did not respond adequately to lithium or valproate disodium. Olanzapine potentiated the antimanic effects of the original treatment but also increased the incidence of adverse effects. (6) In patients with acute mania, the main adverse effects of olanzapine are drowsiness, weight gain, dizziness, and dry mouth. In the trial comparing olanzapine with haloperidol, olanzapine caused fewer extrapyramidal side effects but more weight gain than haloperidol. (7) Olanzapine costs 20 times more than haloperidol in France. (8) In practice, olanzapine is just another neuroleptic approved for the treatment of acute mania in patients with psychotic symptoms and agitation. There is no evidence that olanzapine has the best risk-benefit ratio in this category.