“It is essential to make contact with a hospital laboratory as early as possible.”

The workshop on “Diagnostics in Antimicrobial Resistance: Pathways from Basic Science to Diagnostic Laboratories” delivered plenty of valuable insights and a fair amount of “salt and pepper”, as one speaker put it. It was held on 28 March as part of the NRP 72 programme conference (see box below for the programme and a list of speakers).

The aim was to highlight the routes by which new test procedures developed by researchers make their way into routine diagnostics while taking on board a variety of viewpoints. In this interview, Markus Seeger, the NRP 72 researcher who led the workshop, reflects on the key discussion points.

Interview

Everyone agrees that medicine needs new tests that can detect antibiotic-resistant pathogens faster. At the workshop, however, we heard about a few cases where such tests did not catch on in everyday practice despite being reliable. What kind of things go wrong, Mr Seeger?

To put it simply, there are always other priorities and lots of constraints in real-world practice. For example, the price of new tests is certainly a major factor. Furthermore, if a new test doesn’t slot neatly into a hospital laboratory’s complex working routine or between the laboratory and patients’ doctors, it hasn’t a chance of gaining a foothold in the marketplace. As Patrice Nordmann explained at the workshop, new tests not only have to generate significantly more added value for doctors and labs than conventional tests, the additional work they create for staff must be minimal.

In other words, the cheaper and easier to use, the better?

Both are certainly beneficial. That being said, there are always technologies that medical practitioners are completely unfamiliar with, but which nevertheless promise such clear benefits that the extra work involved in implementing them is worthwhile. For instance, MALDI-TOF, which detects pathogens by mass spectrometry, has become established in most diagnostic laboratories in Switzerland over the past ten years. NRP 72 has both. There are projects that extend firmly established methods, for example by gearing them to other pathogens. Above all though, there are a lot of novel approaches in which technologies – some of which are completely new – are being used in medical applications for the very first time.

One of these examples was presented in greater detail at the workshop – using nanomechanical sensors to determine the response of individual bacteria to antibiotics by means of their movement patterns.

That’s definitely a very encouraging and interesting example. An academic project has spawned a start-up company that has translated the basic science into a product and already come a very long way. Much of what Danuta Cichoka – who as CEO is in charge of the market launch – said could be applied to the implementation of new technology in general.

You are also developing innovative test methods yourself. What key messages have you taken away?

That we need to give our various partners in the real-world environment a very specific explanation of the tangible benefits of our test. And that means focusing clearly on a narrowly defined medical issue beforehand. Clinical practitioners need rapid diagnostic tests primarily when they are faced with a life-threatening infection such as sepsis. In such cases, the most important question a test has to answer is not “what’s the pathogen?” but “which antibiotic will kill it off?”

What other partners in the real-world environment are important in addition to clinicians?

As of a certain point, potential investors who are willing to finance the development process from idea to fully fledged test. Investors generally base their decisions on market analyses and business plans. You have to be able to tell them exactly how – and when – a new test will start generating profits for its producer. But the most important thing is possibly getting a university diagnostic lab with a strong network on board.

Why?

Because ultimately, it’s the labs who work with the tests. Not only do they have technical skills, they are also completely familiar with real-world work procedures. Everyone at the workshop agreed that it was essential to make contact with a hospital lab as early as possible, because taking certain decisions in good time can make a crucial difference to clinical practice and the market. And, as Adrian Egli, who manages a lab himself, as well as developing tests, explained so nicely, the technology doesn’t even have to be fully fledged. Quite the contrary in fact: anyone far-sighted enough to sit down with real-world partners at an early stage has an opportunity to input their needs and expertise into the development process.

And they remain important later on, because once a diagnostic tool has reached market readiness, it has to demonstrate its reliability in real-life clinical testing.

To obtain market authorisation for a test, you have to be able to prove that it is validated, and the requirements for European CE marking, for example, are set to become more stringent over the next few years. That makes clinical labs indispensable partners at this stage too. Already having a good partnership definitely gives you an advantage, because validating a new test means a laboratory has to use it in parallel to the current standard method for a longer period and compare the results. That’s very time-consuming.

But even when a new test method has been certified, that doesn’t mean that hospitals can invoice health insurers for the cost of using it.

That’s an important point, and one that was put primarily to the representative of the Federal Office of Public Health (FOPH) at the workshop. It often takes the FOPH years to add a new diagnostic tool to the list of services reimbursed by health insurers. That’s too long, and it makes it less attractive for investors to put money into the development process. On the other hand, it’s obviously a good thing in principle that the authorities first want to know for certain whether a new test really delivers medical benefits and is cost-efficient.

Does that mean more trials?

Yes, but as Stephan Harbarth emphasised at the workshop, the regulatory environment is poorly set up for trials of this kind, which massively increases the bureaucracy involved in organising them. This is another area where action is needed.

So the fact that new tests deliver faster results does not in itself lead to better treatment?

That’s something else the workshop made clear. If a rapid test enables clinicians to prescribe the right antibiotics sooner than they can at present, but their work processes and priorities prevent them from then winding down the initial treatment, there’s very little gain in terms of combating resistance. And at the end of the day, work processes play a key role. Adrian Egli put forward the notion of “brain-to-brain time”, in other words the time that elapses between a result being received in the lab and the clinician’s decision on what treatment to adopt. And Peter Keller, who also manages a clinical lab, gave a vivid explanation of how the time gained by using new tests is sometimes lost as a result of suboptimal communication paths.

So the benefits yielded by faster diagnostics are also largely determined by the setting in which tests are used?

Although faster diagnostics are important, they are just one factor interacting with others, most notably antibiotic stewardship programmes. If both elements interact effectively, major progress can be achieved.