Feedback control over plasma drug concentrations achieves rapid and accurate control over solid-tissue drug concentrations

Abstract

Electrochemical aptamer-based (EAB) sensors enable the continuous, real-time monitoring of drugs and biomarkers in situ in the blood, brain, and peripheral tissues of live subjects. The real-time concentration information produced by these sensors provides unique opportunities to perform closed-loop, feedback-controlled drug delivery, by which the plasma concentration of a drug can be held constant or made to follow a specific, time-varying profile. Motivated by the observation that the site of action of many drugs is the solid tissues and not the blood, here we experimentally confirm that maintaining constant plasma drug concentrations also produces constant concentrations in the interstitial fluid (ISF). Using an intravenous EAB sensor we performed feedback control over the concentration of doxorubicin, an anthracycline chemotherapeutic, in the plasma of live rats. Using a second sensor placed in the subcutaneous space, we find drug concentrations in the ISF rapidly (30-60 min) match and then accurately (RMS deviation of 8-21%) remain at the feedback-controlled plasma concentration, validating the use of feedback-controlled plasma drug concentrations to control drug concentrations in the solid tissues that are the site of drug action. We expanded to pairs of sensors in the ISF, the outputs of the individual sensors track one another with good precision (R ² = 0.95-0.99), confirming that the performance of in vivo EAB sensors matches that of prior, in vitro validation studies. These observations suggest EAB sensors could prove a powerful new approach to the high-precision personalization of drug dosing.

Figure 1

(A) Electrochemical aptamer-based (EAB) sensors are composed of a methylene blue (MB)-modified aptamer that is site-specifically attached to the surface of a gold electrode. Target binding produces a conformational change in this aptamer, altering the rate of electron transfer. (B) The binding-induced change in electron transfer results in an easily detectable change in peak current when the sensor is interrogated using square-wave voltammetry. (C) To perform measurements of doxorubicin in plasma and subcutaneous space, we utilize a doxorubicin aptamer with a K_d of 5 μM. Data was collected in vitro in whole rat blood at 37°C, with the error bars shown on this graph reflecting standard deviations across 8 independently fabricated devices to illustrate reproducibility. Figure created in BioRender.

Figure 2

To perform feedback-control over plasma drug concentrations, we insert an intravenous sensor in the right jugular vein and an indwelling catheter into the left jugular vein and then connect it to a potentiostat for electrochemical interrogation using square-wave voltammetry. The real-time data this produces is used to inform an adaptive feedback control algorithm that adjusts the rate with which a drug pump delivers the drug to rapidly reach and accurately maintain the desired set point. Here we also employed 1 or 2 sensors in the ventral subcutaneous space to measure how rapidly these solid tissues equilibrate with the plasma drug concentration. Figure created in BioRender.

Figure 3. After the initiation of feedback control, drug levels in the subcutaneous ISF rapidly equilibrate with those in the plasma.

(Top Row) Shown are data sets collected from three different animals (the three columns) illustrating the measured concentration in the plasma (red) and in the ISF (blue). In all three cases the drug concentration in the plasma and ISF rapidly rise to the set point. (Middle Row) Shown are the time-varying infusion rates required to reach and maintain the set-point plasma concentration. Note that the maximum infusion rate differs between rats depending on their body weight. (Bottom Row): After drug concentrations in the plasma and ISF have equilibrated (defined here as subcutaneous concentrations reaching and remaining within 15% of the set point for at least 5 min), the two concentrations remain closely similar. To illustrate this, here we present scatter plots illustrating the correlations between the pairs of measured concentrations after equilibration has been reached. The red points represent the average of vein and subcutaneous measurements obtained over this period, and the error bars reflect the standard deviations of each value. The generally larger standard deviations seen for plasma measurements are due to noise arising from the close placement of these sensors to the animal’s heart. Note: we hypothesize that the slight upward drift seen for both the intravenous and subcutaneous measurements in rat 3 may be due to continued, slow leaking of drug from the pump, as there is no valve that separates it from the animal.

Figure 4

We observe good reproducibility between pairs of subcutaneous sensors placed within individual animals. (Top row) To see this, here we deployed a pair of sensors in the ventral subcutaneous space located ~1 cm away from one another. (Bottom rows) Once again, after the subcutaneous space has equilibrated with the plasma (determined when the subcutaneous levels reach within 15% of the target concentration and maintain set point concentration for at least 5 min), the correspondence between the measured plasma and ISF drug concentrations is excellent for both sensors in the pair. Red points represent the average vein and subcutaneous concentration across the duration of the experiment following equilibration. Error bars represent the standard deviation of the in-vein measurements (red) and the subcutaneous measurements (blue). The corresponding infusion rate data is presented in the SI (Fig. S2).

Figure 5

Measurements performed simultaneously at two sites in the subcutaneous space track one another with good precision. Shown are simple linear regressions of the concentration estimates produced by each of the paired subcutaneous sensors presented in Fig. 4; the Pearson correlations between the paired measurements are exceptional: R² = 0.98, F(1, 762) = 33167, p < 0.0001 and R² = 0.95, F(1, 905) = 16472, p < 0.0001) for rats 4 and 5, respectively. This said, we observe mean, systematic deviations between the two sensors of 9±2% (Rat 4) and 13±3% (Rat 5). This presumably occurs due sensor-to-sensor fabrication variation; equivalent levels of deviation have been reported during in the in vitro characterization of similarly hand-made devices.