INSMED – NTM
The company that is developing Arikayce, a preparation of liposomal amikacin inhalation (LAI), for the treatment of NTM had to withdraw its MAA in Europe earlier this year  because data was not convincing enough for regulators to approve the application. Insmed now has to wait until more data become available from a Phase 3 trial which is now underway and called CONVERT.
A publication of the completed Phase 2 trial was just released giving us a glimpse into the complexity of studying NTM . In this relatively small study, 89 patients with treatment-refractory NTM and positive sputum cultures for MAI or M. abscessus were randomly assigned to aerosolized LAI or placebo. The primary efficacy variable was based on a semi-quantitative outcome measure (a categorized reduction in bacterial count); the LAI arm was numerically better for this endpoint but missed the cut-off for statistical proof of superiority, achieving only a P-value of 0.072. If the team had chosen sputum conversion as their primary outcome variable, the study would have been a resounding success for LAI (P=0.006). Too bad!
Arikayce was not particularly well tolerated. The 12-week treatment course was associated with frequent and occasionally significant AEs (dysphonia, cough, exacerbation of bronchiectasis, oral pain and chest discomfort). These are not minor tolerability issues and would be deemed unacceptable in other more trivial indications. However, when treatment with triple systemic antibiotics is unable to control the pulmonary infection, this should be factored in when assessing therapeutic index and value.
MAI Therapy: Start treatment for pulmonary disease with clarithromycin or azithromycin, plus ethambutol and either rifampin or rifabutin. More advanced disease may require the addition of aminoglycosides. From: Mandell – Ref 
LAI treatment really proIvides more than a ray of hope for treatment-refractory MAI patients. Adjuvant LAI was able to improve the microbiology of this chronic infection which had not responded to long-term therapy with a triple antibiotic cocktail. Not a minor feat!
We should not be too concerned about the fact that mortality or CXR findings or other clinical parameters were not improved, or that walking distance improved but PFT did not; we believe enough in the microbiologic response as the driver of long-term benefit to feel confident that mycobacterial eradication will eventually improve outcomes. With bacterial turn-over in mycobacterial infection occurring only once every 24 hours, nothing very dramatic can be expected within 12-weeks. This duration of therapy is the equivalent of a 30-h antibiotic treatment of an E. coli infection, if we adjust for a cell division time of 20 minutes.
We wish there were a way to make this much needed medication available to patients before the end of a lengthy Phase 3 study which will take years to complete. It seems Arikayce carries a modest amount of risk but brings a very tangible benefit for many with this hard-to-treat infection. If MAI patients had an advocacy group like the CF Foundation or the HIV activists of the late 80s/early 90s, Arikayce would have been conditionally approved.
Alternatively, there is still straight non-liposomal amikacin which can be aerosolized and offered to patients as adjunctive therapy. In an observational trial, most patients achieved sputum conversion and benefited clinically with prolonged treatment.
CARDEAS – VAP Therapy
In a Phase 2 trial Cardeas tested inhaled amikacin+fosfomycin (or AFIS) in VAP patients. In these patients AFIS adjunctive therapy did not improve patient outcomes, not even coming close. While high amounts of drug were deposited in the lungs and bacterial load was reduced in bronchial samples, no clinical benefit was observed over placebo (p=0.7); indeed, mortality was numerically higher in the amikacin+fosfomycin arm. The initial CPIS was 5.6 in the AFIS group with a maximum possible point score of 10. It dropped to 5.0 by the end treatment in the AFIS arm and more so in the placebo group.
Although it looks like a well-designed study, we were somewhat surprised that the company chose ‘Change in CPIS from baseline’ as their primary efficacy measure. The CPIS is such an insensitive scoring system that it is really not very suitable for tracking response to short-term interventions . This is akin to using a bathroom scale to determine how much salt to add to your soup, just too coarse of a measuring stick, much too crude.
Only oxygenation status could be considered to be a useful efficacy marker for the purposes of this trial, but temperature, tracheal secretions, WBC or findings on CXR are unlikely to change much during a 10-day treatment course. Remember, in CF trials it is FEV1 that is used as a measure of response. At any rate, it was nice to see that bacterial counts on BAL showed the expected treatment effect with AFIS.
When FDA set out to determine NI margins: they looked for trials showing HESDE. Investigators are well advised to select their outcome measures carefully to make sure that their primary efficacy variable at least has a chance to meet “ESDE”.
The authors provide several valid explanations why the study failed. We agree with them that prior and concomitant use of broad-spectrum antibiotics (meropenem) were major confounders as well.
(ARADIGM – some other time)
VAT, VAP ventilator-associated tracheitis / -pneumonia
LAI liposomal amikacin inhalation
CPIS clinical pulmonary infection score
MAA market authorization application
MAI Mycobacterium avium-intracellulare
NTM non-tuberculous mycobacteria
HESDE historical evidence of sensitivity to drug effects
FEV1 forced expiratory volume, 1st second
BAL broncho-alveolar lavage
 K Olivier. Randomized Trial of Liposomal Amikacin for Inhalation in Nontuberculous Mycobacterial Lung Disease. AJRCCM Articles in Press. Published on 17-October-2016 as 10.1164/rccm.201604-0700OC
 F Gordin. Chapter 253: Mycobacterium avium Complex, In: Mandell (ed) Principles and Practice of Infectious Diseases. 8th edition. 2014
 M Zilverberg. Ventilator-Associated Pneumonia: The Clinical Pulmonary Infection Score as a Surrogate for Diagnostics and Outcome. CID 2010; 51:S131
 K Davies. Aerosolized amikacin for treatment of pulmonary Mycobacterium avium infections: an observational case series. BMC Pulm Med. 2007;7:2