"Threshold-crossing": A Useful Way to Establish the Counterfactual in Clinical Trials?

A central question in the assessment of benefit/harm of new treatments is: how does the average outcome on the new treatment (the factual) compare to the average outcome had patients received no treatment or a different treatment known to be effective (the counterfactual)? Randomized controlled trials (RCTs) are the standard for comparing the factual with the counterfactual. Recent developments necessitate and enable a new way of determining the counterfactual for some new medicines. For select situations, we propose a new framework for evidence generation, which we call "threshold-crossing." This framework leverages the wealth of information that is becoming available from completed RCTs and from real world data sources. Relying on formalized procedures, information gleaned from these data is used to estimate the counterfactual, enabling efficacy assessment of new drugs. We propose future (research) activities to enable "threshold-crossing" for carefully selected products and indications in which RCTs are not feasible.

From adaptive licensing to adaptive pathways: delivering a flexible life-span approach to bring new drugs to patients.

The concept of adaptive licensing (AL) has met with considerable interest. Yet some remain skeptical about its feasibility. Others argue that the focus and name of AL should be broadened. Against this background of ongoing debate, we examine the environmental changes that will likely make adaptive pathways the preferred approach in the future. The key drivers include: growing patient demand for timely access to promising therapies, emerging science leading to fragmentation of treatment populations, rising payer influence on product accessibility, and pressure on pharma/investors to ensure sustainability of drug development. We also discuss a number of environmental changes that will enable an adaptive paradigm. A life-span approach to bringing innovation to patients is expected to help address the perceived access vs. evidence trade-off, help de-risk drug development, and lead to better outcomes for patients.

Interstitial methotrexate kinetics in primary breast cancer lesions.

The transfer of cytotoxic agents across the tumor endothelium into the interstitial tumor space is considered a critical step in clinical resistance of solid tumors to antineoplastic chemotherapy. However, experimental data on drug transfer from the blood into the interstitium of solid tumors are scarce. Therefore, in this study, we used an innovative technique, in vivo microdialysis, for measuring interstitial tumor pharmacokinetics and plasma-to-tumor transfer rates of methotrexate (MTX) in breast cancer patients. Microdialysis probes were inserted into the primary tumor and the periumbilical s.c. adipose layer of nine previously chemotherapy-naive breast cancer patients to monitor interstitial concentrations following i.v. administration of MTX (40 mg/m2) during a three-drug treatment regimen. Mean interstitial MTX load in breast tumors, expressed as area under curve (AUC), was 60 +/- 20% (mean +/- SE; coefficient of variation = 100%) of mean plasma MTX load. There was no correlation between plasma AUC and the AUC in the interstitial space of tumor tissue (P = 0.93). Not one of the parameters plasma, interstitial tumor load, and transfer rate of MTX to the interstitial space was associated with favorable clinical response. In conclusion, plasma levels of MTX are not predictive of intratumor levels. There is a high interindividual variability in transendothelial MTX transfer. Under the present conditions, access of MTX to the interstitial space is not a rate-limiting step for clinical response to chemotherapy.

Cisplatin-induced renal effects and thromboxane A2 receptor blockade.

The aim of this study was to evaluate the renal protective effect of linotroban, a thromboxane A2 receptor antagonist, in 25 patients with malignant tumours scheduled for cisplatin therapy. Cisplatin was administered 1 h after the start of a 24-h continuous infusion of linotroban or placebo. Glomerular filtration rate and effective renal plasma flow were measured. Infusions of cisplatin decreased glomerular filtration rate by 17 +/- 25 mL min-1 (P = 0.049 vs. baseline) and effective renal plasma flow by 94 +/- 150 mL min-1 (P = 0.049 vs. baseline) in the placebo group. In the linotroban group a decrease in glomerular filtration rate by 11 +/- 18 mL min-1 (P = 0.050 vs. baseline) and in effective renal plasma flow by 26 +/- 63 mL min-1 (P = 0.2 vs. baseline) was noted. However, no difference was noted between groups in response to treatment. Our findings indicate that linotroban may not be useful for prevention of cisplatin's acute nephrotoxic effects.

5-fluorouracil kinetics in the interstitial tumor space: clinical response in breast cancer patients.

Several anticancer drugs fail to exhibit sufficient activity against solid tumors in vivo despite effective inhibition of tumor cell growth in vitro. This may be due to impaired drug transfer from plasma into solid tumors. The present study, therefore, aimed at measuring interstitial tumor 5-fluorouracil (5-FU) pharmacokinetics and 5-FU transfer rates from plasma into the tumor interstitium in breast cancer patients. Microdialysis probes were inserted into the primary tumor and the periumbilical s.c. adipose layer of 10 breast cancer patients (8 females and 2 males) scheduled to receive neoadjuvant chemotherapy due to locally advanced breast cancer. Thereafter, patients received 5-FU (600 mg/m2, i.v). 5-FU kinetics were followed in plasma and tumor and s.c. interstitial fluid. Mean interstitial 5-FU load, expressed as area under curve (AUC), in breast tumors was 61 +/- 11% (means +/- SE) of the mean plasma 5-FU load. 5-FU displayed similar kinetics in the interstitial space of s.c. adipose tissue and tumor tissue. A high interstitial tumor AUC was associated with increased tumor response. There was no association with tumor response for s.c. or plasma AUC of 5-FU. Measurement of interstitial drug concentrations in breast tumors by in vivo microdialysis may predict response to chemotherapy. This information may explain drug resistance in some patients and help to optimize dosing and administration schedules. In the future, selection of novel cytotoxic compounds with favorable tumor penetration characteristics may become possible.

First-pass uptake of methotrexate after intra-arterial bolus injection in human head and neck cancers.

The aim of this study was to quantify first-pass uptake of methotrexate into tumour and healthy tissue after intra-arterial (i.a.) injection in patients with head and neck cancers. Twenty-one patients were studied; during surgery, a bolus injection of 10 mg of 3H-labelled methotrexate was administered into the tumour-supplying artery. At 15-20 s after injection, and again 2-3 h later, a biopsy of the tumour and of surrounding healthy tissue was taken and radioactivity was measured. In 8 of 17 pretreated patients the biopsy specimens contained no tumour cells. In tumour-bearing patients, radioactivity in tumour was three times higher than in healthy tissue (P = 0.006). In tumour-negative patients there was no difference in activity between the two biopsy sites (P = 0.889). However, after 2-3 h the concentration of methotrexate in all surgical specimens returned to near background level and the initial difference between tumour-positive and tumour-negative samples was lost. Our results show that i.a. injection of methotrexate does not necessarily lead to high tumour cell exposure.

True epineural neurolysis in Morton's neuroma: a 5-year follow up.

Morton's neuroma is a frequent cause of metatarsalgia. Neuroma resection was the usual recommended surgical treatment. Failure rate of neurectomy can be as high as 14% to 21%, and treatment of recurrences is difficult. The authors have treated Morton's neuroma by neurolysis since 1985. They present their results in a group of 40 patients with 5 years follow up. Thirty-seven of 40 patients had an excellent result after neurolysis and 35 patients had normal toe sensitivity at the date of examination. Thirty-nine patients stated they would undergo the operation again if necessary.

Measurement of extracellular fluid carboplatin kinetics in melanoma metastases with microdialysis.

Clinical anti-tumour efficacy of anti-cancer drugs is a function of dose intensity, i.e. the concentration--time profile in tumour tissue. Hence, information on drug concentration profiles in tumours is of critical importance but appropriate methods for measurement are lacking. The aim of the present study was to obtain, by microdialysis sampling, concentration--time profiles in a solid tumour (melanoma) of a model anti-cancer drug, carboplatin, and thereby to assess the scope of microdialysis for tumour pharmacokinetic studies in man. Six patients with cutaneous melanoma metastases at the extremities or body trunk, scheduled to receive carboplatin (400 mg m-2 i.v.) were studied. Carboplatin concentrations were monitored in serum, intratumoral and subcutaneous tissue. Calibration of the microdialysis probes was carried out in vitro and in vivo with use of the retrodialysis method. Complete carboplatin concentration vs time profiles in tumour and subcutaneous tissue were obtained. Major pharmacokinetic parameters (maximum concentration, time to maximum concentration, area under the curve, elimination half-life) were calculated for tissues and tumour/serum concentration ratios for carboplatin were derived. Mean free concentrations of carboplatin in cutaneous melanoma metastases reached only about 50-60% of total serum levels; maximal intratumoral concentrations were 7.6 (+/-2.0; s.e.m.) microgram/ml, mean concentrations in subcutaneous tissue were similar to those in tumour. The present study demonstrates that microdialysis is a novel tool for measuring drug concentrations in solid tumours in humans in vivo and appears to be a valuable addition for pharmacokinetic/pharmacodynamic studies in oncology.

Effects of cisplatin on urinary thromboxane B2 excretion.

PURPOSE: Thromboxane A2 (TxA2) is implicated in the pathogenesis of various forms of drug-induced renal damage. Based on previous functional studies, we postulated that cis-dichlorodiammineplatinum (cisplatin) induces intrarenal TxA2 synthesis. To test this hypothesis, we measured urinary excretion of thromboxane B2 (TxB2), the stable inactive metabolite of TxA2, during and after cisplatin administration. PATIENTS AND METHODS: The study included 16 patients with malignant disease who were scheduled to receive cisplatin (100 mg/m2) and 11 healthy subjects who received the same amount of fluid loading and the same concomitant medication as the patients but no cisplatin. Total urine output was collected in seven intervals from 24 hours before until 72 hours after the start of prehydration. Urinary immunoreactive TxB2 was measured. RESULTS: There was a marked increase (4.5 +/- 1.6-fold; mean +/- SEM) in urinary TxB2 excretion in patients during and immediately after cisplatin infusion. This increase was significant compared with baseline and the control group. CONCLUSION: High-dose cisplatin causes an acute increase in urinary excretion of TxB2. This likely represents enhanced intrarenal synthesis of TxA2, in response to an acute damaging effect of cisplatin on the kidneys. These findings warrant further studies to evaluate the renoprotective effect of anti-TxA2 intervention in patients receiving high-dose cisplatin.

Studies on the effect of two angiotensin-converting enzyme inhibitors, captopril and cilazapril, on platelet and vascular prostaglandin metabolism in vivo.

We have studied in 12 healthy volunteers the effects of two angiotensin-converting enzyme (ACE) inhibitors, captopril and cilazapril, on vascular and platelet prostaglandin metabolism, in a double-blind, placebo-controlled, randomized cross-over study. Formation of 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha) and thromboxane B2 (TxB2) was measured locally at the site of a microvascular injury. Similar amounts of TxB2 and 6-keto-PGF1 alpha were generated following administration of either ACE inhibitor as compared to placebo. It is concluded that neither captopril nor cilazapril significantly influence vascular and platelet prostaglandin metabolism.

Endothelin-induced venoconstriction is unaffected by type 2-diabetes: in vivo effect of histamine on the endothelin action on veins.

In vitro studies indicate that the intact endothelium prevents the access of circulating endothelin to vascular smooth muscle cells. Disease states like diabetes, that are associated with endothelium dysfunction, might facilitate the access of endothelin to vascular smooth muscle, causing increased vasoconstriction. To investigate whether differences between diabetic and nondiabetic subjects could be detected on superficial veins, venous diameter changes in response to local infusions of endothelin (1.53, 3.11 and 6.22 pmol/min) were compared in type 2-diabetics with albuminuria (greater than 300 mg/day) and borderline hypertension (diastolic blood pressure 90-95 mmHg), and in healthy volunteers. In addition, we studied the effect of histamine on the endothelin-induced venoconstriction. The dorsal hand vein compliance technique was employed. Endothelin caused a dose-dependent, slow-onset constriction of the hand vein in all subjects, without any difference in endothelin responsiveness between diabetic and control subjects. The maximal venoconstriction at 6.22 pmol endothelin/min was 29 +/- 26% (% of venous diameter at base line) in diabetics and 23 +/- 22% in controls (p greater than 0.02). Co-infusion of endothelin and a dose of histamine with minimal venodilatory effect ("ED5-ED20") had no influence on the endothelin responsiveness of the hand veins, in that the maximal venoconstriction after 6.22 pmol endothelin/min was 25 +/- 23% without, and 23 +/- 16% with added histamine. Submaximally venodilating histamine doses ("ED60-ED90") markedly attenuated the endothelin-induced venoconstriction (maximal venoconstriction: 63 +/- 43%). These data, obtained in veins, argue against a generalized defect in vascular responsiveness to endothelin associated with diabetes.

Responsiveness of superficial hand veins to alpha-adrenoceptor agonists in insulin-dependent diabetic patients.

1. Diabetic autonomic neuropathy causes loss of sympathetic cardiovascular control and is associated with increased vascular sensitivity to catecholamines. Supersensitivity to catecholamines could be due to either a postsynaptic increase in vascular sensitivity or to decreased catecholamine uptake into peripheral sympathetic nerve endings. 2. To differentiate between these possible mechanisms we have measured the responsiveness in vivo to noradrenaline and phenylephrine with local infusions into peripheral veins of diabetic patients with and without symptomatic autonomic neuropathy and of healthy control subjects. The dorsal hand vein compliance technique was used. 3. Symptomatic diabetic patients required significantly lower doses of noradrenaline for half-maximal venoconstriction (ED50) (geometric mean 2.14 ng/min) than control subjects (geometric mean 6.61 ng/min, P = 0.032), but there was no difference in the results from the phenylephrine dose-response curves between the groups. There were no differences in venous responsiveness to noradrenaline or phenylephrine between the asymptomatic diabetic group and the control group. However, in the asymptomatic diabetic group, postural blood pressure change (an index of loss of sympathetic control) was correlated with the ED50 for noradrenaline (r = 0.74, P = 0.014), but not with the ED50 for phenylephrine. In the control group the ED50 values for noradrenaline and phenylephrine were correlated with each other (r = 0.81, P = 0.0005). 4. Both vasopressor drugs act on vascular alpha-adrenoceptors, but only noradrenaline is taken up into peripheral sympathetic nerve endings. Our results suggest that, in diabetic patients, vascular supersensitivity to catecholamines is primarily determined by decreased neuronal catecholamine uptake. A postsynaptic increase in vascular alpha-adrenoceptor stimulation does not appear to be prominent in diabetic autonomic neuropathy.

In vivo studies on alpha-adrenergic receptor subtypes in human veins.

We studied in vivo responsiveness of venous alpha 1- and alpha 2-adrenoceptors, measuring the diameter changes in superficial veins in response to alpha-adrenergic agonists and antagonists in healthy human volunteers. The dorsal hand vein technique was used because it permits complete dose-response studies of venous constriction without confounding reflex alterations. Local infusions of all agonists studied induced dose-dependent contraction of the hand vein; the maximal effects (Emax) were: norepinephrine (88% +/- 10%), methoxamine (97% +/- 5%), phenylephrine (95% +/- 6%), clonidine 54% +/- 12%), and azepexole (68% +/- 26%). Clonidine reduced the norepinephrine-induced venoconstriction by 11% +/- 10%. Oral doses of 1 mg prazosin antagonized the venoconstriction induced by norepinephrine, methoxamine, and clonidine, but not by azepexole. Yohimbine-antagonism was observed against all agonists studied. Inhibition by yohimbine of clonidine-induced venoconstriction was irreversible over 60-180 min. Results show that the in vivo effects on veins of alpha-adrenergic agonists are in good agreement with results from in vitro experiments. Agonists with alpha 1- and alpha 2-adrenoceptor subtype selectivity cause venoconstriction in vivo, but alpha 2-receptor mediated constriction is intrinsically weaker. Clonidine acts as a partial antagonist against norepinephrine, presumably on postsynaptic alpha 2-receptors. At high doses, alpha 2-adrenoceptor subtype selectivity of clonidine and yohimbine appear to be partially lost in vivo.

Bioequivalence between two furosemide-spironolactone formulations: a pharmacokinetic and pharmacodynamic approach.

Two formulations of the drug combination furosemide (20 mg)-spironolactone (100 mg) were tested for bioequivalence in a randomized crossover trial in 12 healthy volunteers. By comparing the AUC values derived from drug serum concentrations, bioequivalence was only achieved for canrenone, the main metabolite of spironolactone, but not for furosemide. A significant difference in the tmax values indicates sustained release of furosemide from one of the formulations. By contrast, bioequivalence was achieved if pharmacodynamic criteria, such as urine volume and Na+ and Cl- excretion over a period of 12 hours, were used. Fractional measurements of urinary volume and electrolyte excretion (0 to 3 h, 3 to 6 h, 6 to 12 h) correlated with the different tmax values for both formulations. These data indicate that bioequivalence is more conclusively verified on the basis of pharmacodynamic parameters than on the basis of pharmacokinetic parameters. These considerations are applicable in particular to drugs displaying large inter-individual variations in serum levels and/or a poor correlation between serum levels and effect.

Primary defect in alpha-adrenergic responsiveness in patients with varicose veins.

Responsiveness of superficial hand veins to local infusions of noradrenaline was compared in patients with primary varicose veins and in healthy volunteers by use of the dorsal hand vein technique. Patients with varicose veins required significantly higher doses of noradrenaline for half-maximal venoconstriction than the dose required by control subjects (geometric mean, 11.6 ng/min in patients compared with 4.8 ng/min in control subjects; p = 0.006). Noradrenaline responsiveness in varicose veins was not significantly different from hand vein responsiveness in the same patients. Our findings indicate a constitutional decrease in venous alpha-adrenergic receptor responsiveness in patients with varicosities. Dilation of varicose veins does not further affect noradrenaline-induced venoconstriction.

Normal responsiveness of superficial hand veins to alpha- and beta-adrenergic stimuli in allergic asthma: effects of terbutaline and prednisolone on beta-adrenergic responsiveness.

Impaired function of the adrenergic-receptor system has been postulated to contribute to the pathogenesis of bronchial asthma. Using the dorsal hand-vein compliance technique, we compared the changes in diameter of superficial hand veins in response to phenylephrine, an alpha-adrenoceptor agonist, and to isoproterenol, a beta-adrenoceptor agonist, in 14 untreated patients with allergic asthma and in 16 nonatopic control subjects. There were no significant differences in the median effective dose of phenylephrine that produced 50% of maximal venoconstriction (ED50) or in the maximal response (Emax) between the two groups. Bronchial hyperreactivity (assessed by methacholine-challenge tests) in the patients with asthma was uncorrelated with the ED50 or Emax of isoproterenol. These results demonstrate no evidence for a generalized change in alpha- or beta-adrenergic responsiveness on smooth muscle cells in asthma. Hand-vein responsiveness to isoproterenol was unchanged after treatment for 7 days with oral terbutaline (5 mg three times per day). Thus, unlike leukocytes, smooth muscle appears not readily susceptible to beta-adrenoceptor desensitization in vivo. Local infusions of prednisolone or dexamethasone during 2 hours and systemic administration of dexamethasone (24 hours) caused a significant fall in the Emax for isoproterenol. The mechanism of attenuation of beta-adrenoceptor responsiveness by corticosteroids remains to be determined.

Effect of ritanserin, a 5-hydroxytryptamine2-receptor antagonist, on platelet function and thrombin generation at the site of plug formation in vivo.

To investigate the role of serotonin in platelet plug formation we studied, in eight healthy volunteers, the effect of ritanserin (a 5-hydroxytryptamine2-receptor antagonist) on the platelet release reaction (represented by beta-thromboglobulin release) platelet prostaglandin metabolism (represented by thromboxane B2 formation), and thrombin generation (represented by fibrinopeptide A formation) in the microvasculature. After administration of ritanserin lower amounts of thromboxane B1 were generated in the initial stages of plug formation, suggesting an inhibitory effect on the platelet prostaglandin metabolism. Similar amounts of beta-thromboglobulin were released after the administration of ritanserin compared with placebo, indicating a minor effect of ritanserin on the release reaction. Reduction of thrombin formation by ritanserin in the later stages of hemostasis suggested an inhibitory effect of this substance on the procoagulatory activity of platelets or endothelial cells. This could be attributable to interference with the formation or function of coagulation factor complexes on cell surfaces, or it could be the consequence of a reduction of the platelet activity.

Cilazapril and enalapril inhibit local angiotensin I conversion in human veins but lack direct venodilating properties.

We studied the angiotensin-converting enzyme (ACE)-dependent and ACE-independent (direct) effects of two ACE inhibitors, cilazaprilat and enalaprilat, in 12 healthy human subjects. The dorsal hand vein compliance technique was used because venous constriction and relaxation, independent of reflex responses and systemic ACE inhibition, can be measured by local infusions of very small amounts of drugs. Angiotensin I (dose range 6-1,550 ng/min) was infused alone or coinfused with cilazaprilat or enalaprilat (dose range 7.8-3,900 ng/min). In separate experiments, cilazaprilat or enalaprilat (dose range 3.9-31 micrograms/min), or prostaglandin I2 (PGI2, dose range 0.13-32 ng/min) was infused into veins that had been submaximally preconstricted with phenylephrine. Angiotensin I caused a marked venoconstriction limited by rapid tachyphylaxis. At doses greater than 78 ng/min, cilazaprilat and enalaprilat completely inhibited angiotensin I-induced venoconstriction. This inhibition was reversible after 14-31 min, suggesting an inhibition of ACE associated with the vein wall. Infusions of cilazaprilat or enalaprilat had no effect on the diameter of the vein at rest or after submaximal preconstriction with phenylephrine. In contrast, exogenous PGI2 was a potent venodilator in our system. We conclude that cilazaprilat and enalaprilat are inhibitors of ACE associated with the vein wall, but there is no evidence for either drug of direct, ACE-independent, prostaglandin-mediated vasodilation.