Pathogens persist in structure, and surveillance is learning to see it

Measles moves regionally as vaccination rises

As vaccination coverage increases, measles does not simply fade. It changes shape. A preprint posted October 26 introduces a scale-free metric called roundness to distinguish local persistence from persistence driven by regional metapopulation dynamics, using subnational incidence in Nigeria. The finding is that vaccination reduces local prevalence, and as it does, colonization and extinction events across the regional network come to dominate persistence. In Nigeria, roundness correlates with vaccine coverage at r = 0.70, which supports the metric as a practical signal rather than a theoretical one. m:…4784

For public health planning, the distinction changes the intervention. Locally sustained transmission calls for local coverage; regionally driven persistence calls for coordination across the network and attention to importation. A metric that tells the two apart lets officials choose the right response instead of applying one playbook everywhere. m:…4784

The Nigeria result gives the metric practical weight. Where vaccine coverage is higher, roundness behaves as the theory predicts, with a correlation of 0.70. That is not a perfect relationship, and the confidence interval leaves room for local variation, but it is strong enough to use: a region whose persistence looks metapopulation-driven should be investigated as a network problem, with attention to movement and importation, rather than treated as a local coverage failure. m:…4784

One hospital, few drugs left

A second preprint reports surveillance from a tertiary oncology center between 2022 and 2026, and the numbers are stark. Carbapenem-resistant Klebsiella pneumoniae was the dominant phenotype, peaking at 76.4% in 2024, and aggregate multidrug resistance rates of 76-87% run 2-3 times higher than the global pooled mean of 28.7%. Amikacin resistance rose from 8.5% in 2022 to 38.1% in early 2026, reaching 60% in intensive care units, leaving only polymyxins and tigecycline with retained activity at the center. m:…1310

Without local surveillance, stewardship is guesswork. With it, the center can see exactly which agents still work and how fast the rest are failing. m:…1310

The pattern is not static, either. By 2025-2026, resistant E. coli had emerged as the dominant pathogen in the center's haematological and surgical departments, showing that the resistance problem shifts between organisms as well as between drugs. Surveillance that tracks only one organism will miss the next one. The center's data argues for watching the whole ecology, not just the worst actor of the moment. m:…1310

One host, several strains

The third preprint attacks a different blind spot: what lives inside a single patient. A culture-free sequencing method achieves thousand-fold enrichment of H. pylori DNA and recovers near-complete genomes from gastric biopsies, bypassing the distortion that culture introduces. It resolves resistance-associated variants for six commonly used antibiotics with greater accuracy than PCR-based and probe-based commercial panels, and it reveals that individual hosts may carry multiple co-existing strains or variants with different oncogenic potential. m:…3621

That last point changes treatment logic. If one host carries several strains with different resistance and cancer-risk profiles, a single-strain test can mislead. The method points toward genome-guided eradication. Across all three papers, the theme holds: persistence is structural, and the tools that reveal the structure are what make it manageable. The throughline is practical, not theoretical: persistence has a shape, and once surveillance can see the shape, the response can match it. m:…3621 m:…4784