Joint targets for beta-lactam / beta-lactamase inhibitor pairs
A beta-lactam paired with a beta-lactamase inhibitor is not one medicine with a long name. One partner usually does the killing work. The other protects that work from enzymes. If protection fails, the beta-lactam can look adequate on a plot and still lose in the patient.
That is why combination PK/PD often needs a joint rule: both conditions true in the same virtual subject at the same evaluation window.
Simulation logic for combinations, with ceftazidime-avibactam style problems as a teaching frame. Not product-specific clinical advice.
Two conditions, one success flag
Suppose drug A must meet a free-exposure target against MIC, and drug B must stay above a threshold that preserves A’s activity. For each simulated person:
[ = {A} {B} ]
Joint PTA is the fraction of people for whom that intersection is true. It is not the average of PTA(A) and PTA(B). It is not “A looks fine, so ignore B.”
Why correlation between clearances matters
If A and B clearances move together, the joint tail can be very different from two independent draws. Ignoring correlation can make joint failure look rarer or more common than your evidence supports. State the correlation assumption.Mistakes that look sophisticated
- Simulating only the beta-lactam because “that is the antibiotic.”
- Reporting the mean of two PTAs as if it were joint attainment.
- Borrowing an inhibitor threshold from another matrix or population without labeling.
- Collapsing both drugs into one fake combined concentration.
Simulate free concentration-time for A and for B in each virtual subject. Keep units and binding honest for each component.
Apply each target. Mark success only when both pass. Build PTA versus MIC from that joint indicator. Optionally weight by MIC ecology for joint CFR.
Average(PTA_A, PTA_B), or "A passed so count success." Those shortcuts answer different questions and can hide inhibitor-limited regimens.
Why this shows up in CAZ-AVI style work
Ceftazidime and avibactam are a teaching example because the beta-lactam exposure and the inhibitor exposure are not interchangeable. A regimen can clear a ceftazidime rule while avibactam is the limiting partner, or the reverse depending on the scenario. Component-level evidence reconstruction exists partly because papers report pieces of that system separately, and the modeler has to keep those pieces labeled until a joint rule is applied on purpose.
A tiny counting example
Imagine 100 virtual subjects.
- 90 pass the beta-lactam rule.
- 80 pass the inhibitor rule.
- Only 70 pass both.
Joint PTA is 70%. The average of 90% and 80% is 85%. That 15-point gap is exactly why intersection logic exists. If clearances are positively correlated, joint failures may cluster; if not, the overlap changes again. Either way, report the joint indicator you actually coded.
When you write results, show component PTA only as supporting context. Lead with joint PTA if the clinical object is the combination.
If you need a debugging view, plot the fraction failing A only, B only, and both. That split tells you whether the regimen is beta-lactam limited, inhibitor limited, or jointly fragile. A single joint number hides that diagnosis.
Takeaway
For BL/BLI pairs, protection failure is regimen failure. Joint PTA is an intersection inside each simulated person. Build it that way, then be transparent about thresholds, correlation, and which MIC the beta-lactam rule refers to.
Previous: Intermittent vs continuous infusion · Next: Uncertainty and provenance · Case study: CAZ-AVI end-to-end.