REVIEW
Feedback Medicine: Control Systems Concepts in Personalised, Predictive Medicine and Combinatorial Intervention
Tissues and Organs · Submitted 27 Mar 2006 · 19 references
Pith's verdict: Honest working notes with sound algebra, but the headline failure-rate benefit depends on an unexamined independence assumption that plausibly collapses under correlated drug sensitivities.
T0 · AI review
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
Feedback control theory predicts that equal-dose drug combinations reduce treatment failure and side effects, while overriding the body's own regulatory loops causes hidden drift.
The reading
What carries the argument
The load-bearing object is the closed feedback loop formed by patient, measurement, and treatment controller, analysed with linear discrete-time models: a single-state example $x_{k+1}=s_A u_{Ak}+s_B u_{Bk}$ for the drug-combination result, and a three-variable static model $y = M_u u + x$ with integral homeostatic loops $\dot{x}_i = r_{bi} - y_i$ for the loop-interaction result. The key identity is that coordinated equal-dose treatment replaces the patient's sensitivity to one drug by the average sensitivity to two, which narrows the distribution of responses across a population. The key control-theoretic mechanism is the rank and transmission-zero condition: a plant that already contains an internal perfect-adaptation (integral) loop for a particular variable has a transmission zero at that variable, so adding an external integral loop for the same variable leaves a marginally stable hidden mode that drifts until saturation.
Extended reading notes
Core claim
On the paper's own terms, feedback medicine is the application of linear feedback theory to the administration of treatment, with the patient modelled as a dynamical system whose measured outputs are diagnostic indicators and whose inputs are drug components. The first discovery is a variance-reduction argument: for a patient whose next indicator value is $x_{k+1} = s_A u_{Ak} + s_B u_{Bk}$, treating with equal doses of two drugs replaces the single sensitivity $s_A$ by the average $\frac{1}{2}(s_A+s_B)$; because averaging lowers the spread of patient sensitivities, the failure rate in the worked example falls from 5% to 1%, and the four-cycle success rate rises from 87% to 95%, while equal-dose treatment also minimises peak and RMS drug dose. The second discovery is a loop-interaction warning: if the body already has an internal integral feedback loop that perfectly adapts a variable, closing an external integral loop around the same variable creates a transmission zero and hence marginally stable hidden drifting modes, so external perfect adaptation should be reserved for variables whose internal regulation is inoperative. The accompanying simulations show that regulating only the broken variable, with equal control weights, avoids saturation and large transients that appear when all three variables are externally regulated to slightly wrong reference values.
Load-bearing premise
The worked benefits assume that a patient's response to drugs is well approximated by static linear gains (e.g. $x_{k+1} = s_A u_{Ak} + s_B u_{Bk}$) plus integral homeostatic loops; if real regulation is strongly nonlinear or time-varying, the predicted reductions in failure and the warning about fighting loops may not carry over.
Editorial extensions
If this is right
- Coordinated equal-dose combination of two similarly effective drugs should be a default candidate in personalised treatment design, because it reduces both failure probability and side-effect peaks without requiring prior knowledge of which drug suits the patient.
- Treatment regimes should be evaluated on transient behaviour, not just steady-state efficacy; aggressive one-step-ahead dosing creates large interacting transients that slow, gentler adjustment dampens.
- External controllers should avoid closing perfect-adaptation loops on variables the body already regulates perfectly; otherwise hidden drift will eventually saturate internal or external regulators.
- Population-level variability, not just average response, is the quantity that determines whether a combination strategy helps; averaging sensitivities across drugs shrinks the effective population spread.
- Sampling of diagnostic indicators should respect the dynamics of the disease and treatment, since spot samples can alias and miss intersample excursions.
Reading between the lines
- The sensitivity-averaging argument extends beyond two drugs: splitting a dose across more independent drugs should further shrink the spread of patient response, though the marginal benefit diminishes and toxicity interactions could reverse the advantage.
- The hidden-drift warning suggests a diagnostic test: look for slow, largely invisible drift in medication levels or internal regulators of patients on long-term therapy, since it signals that an external loop is fighting an internal one.
- A testable extension would randomise patients between single-drug and equal-dose combination therapy while measuring individual drug sensitivities; the model predicts the failure-rate gap should track the variance of the average sensitivity, not the means.
- The multivariable framing implies that drugs with strong off-target interactions could be used constructively if their interaction matrix is known, turning coupling from a problem into a design variable.
Original abstract
In its broadest definition, systems biology is the application of a `systems' way of thinking about and doing cell biology. By implication, this also invites us to consider a systems approach in the context of medicine and the treatment of disease. In particular, the idea that systems biology can form the basis of a personalised, predictive medicine will require that much closer attention is paid to the analytic properties of the feedback loops which will be set up by a personalised approach to healthcare. To emphasize the role that feedback theory will play in understanding personalised medicine, we use the term feedback medicine to describe the issues outlined.In these notes we consider feedback and control systems concepts applied to two important themes in medical systems biology - personalised medicine and combinatorial intervention. In particular, we formulate a feedback control interpretation for the administration of medicine, and relate them to various forms of medical treatment.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
In its broadest definition, systems biology is the application of a `systems' way of thinking about and doing cell biology. By implication, this also invites us to consider a systems approach in the context of medicine and the treatment of disease. In particular, the idea that systems biology can form the basis of a personalised, predictive medicine will require that much closer attention is paid to the analytic properties of the feedback loops which will be set up by a personalised approach to healthcare. To emphasize the role that feedback theory will play in understanding personalised medicine, we use the term feedback medicine to describe the issues outlined.In these notes we consider feedback and control systems concepts applied to two important themes in medical systems biology - personalised medicine and combinatorial intervention. In particular, we formulate a feedback control interpretation for the administration of medicine, and relate them to various forms of medical treatment.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | In its broadest definition, systems biology is the application of a `systems' way of thinking about and doing cell biology. | background 1.00 | 0.05 | 0.03 | asserted 1.00 |
| s1 | By implication, this also invites us to consider a systems approach in the context of medicine and the treatment of disease. | background 0.92 | 0.07 | 0.05 | hedged possibility 0.68 |
| s2 | In particular, the idea that systems biology can form the basis of a personalised, predictive medicine will require that much closer attention is paid to the analytic properties of the feedback loops which will be set up by a personalised approach to healthcare. | background 0.71 | 0.07 | 0.25 | asserted 0.56 |
| s3 | To emphasize the role that feedback theory will play in understanding personalised medicine, we use the term feedback medicine to describe the issues outlined.In these notes we consider feedback and control systems concepts applied to two important themes in medical systems biology - personalised medicine and combinatorial intervention. | contribution 0.91 | 0.60 | 0.14 | asserted 0.93 |
| s4 | In particular, we formulate a feedback control interpretation for the administration of medicine, and relate them to various forms of medical treatment. | contribution 0.95 | 0.85 | 0.14 | asserted 0.99 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Figures reproduced from arXiv:q-bio/0603032 by the authors.
References
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.