In-country cerebrospinal fluid metagenomic next generation sequencing for pathogen detection in Lima, Peru
Background: Central nervous system (CNS) infections are a major cause of morbidity and mortality in low- and middle-income countries such as Peru. However, their epidemiology is poorly characterized due to a wide range of endemic and opportunistic pathogens involved and limited access to specialized neurodiagnostic testing. Cerebrospinal fluid (CSF) metagenomic next-generation sequencing (mNGS) can detect any pathogen in a single hypothesis-free assay but has largely been confined to reference laboratories in high-income countries due to high cost and infrastructure requirements. We piloted an in-country, research-based CSF mNGS protocol in Lima, Peru, as a proof-of-concept study to assess the feasibility of implementing this approach in a resource-limited setting. Methods: Adults admitted with suspected CNS infection at Hospital Nacional Dos de Mayo, a large public hospital in Lima, were enrolled from August 2024 to April 2025, and approximately 0.5-2 mL of surplus CSF was collected. CSF mNGS (DNA and RNA) was performed in-country on a benchtop sequencer with half-volume library reactions and cloud-based analysis (CZID). Paired sequencing at a significantly higher depth was performed on a subset of samples in the United States for validation. Results: Among 43 participants (median age 48 years; 21% living with HIV), mNGS performed in Peru detected 20 pathogens in 18 samples (42%), of which 65% were not identified by routine clinical testing. Nine of 15 patients treated empirically for tuberculous meningitis were found to have a non-mycobacterial infection. Detected organisms included Toxoplasma gondii, Taenia solium, Treponema pallidum, human T-lymphotropic virus 1, herpesviruses, and, on deeper sequencing, Venezuelan equine encephalitis virus. Deep sequencing yielded 3.7-fold (DNA) and 6.9-fold (RNA) more reads and recovered four pathogens missed at shallower-depth. Conclusions: In-country CSF mNGS was feasible in Lima, Peru and recovered a clinically and epidemiologically meaningful spectrum of CNS pathogens, most missed by routine testing. By utilizing available, lower-output sequencers, we overcame a principal barrier to deploying this technology where the burden of undiagnosed CNS infection is greatest.
Authors
- Kelsey C. Zorn (ORCID: https://orcid.org/0000-0003-1227-2137)
- Chloe Gerungan
- Andrés G. Lescano (ORCID: https://orcid.org/0000-0001-9779-633X)
- Renzo Villanueva-Villegas (ORCID: https://orcid.org/0000-0002-7820-0016)
- Carolina Cucho-Espinoza (ORCID: https://orcid.org/0000-0003-3529-4830)
- Meddly L. Santolalla (ORCID: https://orcid.org/0000-0003-4808-2896)
- Pablo Tsukayama (ORCID: https://orcid.org/0000-0002-1669-2553)
- Felicia C. Chow (ORCID: https://orcid.org/0000-0001-9440-9349)
- Félix Llanos-Tejada (ORCID: https://orcid.org/0000-0003-1834-1287)
- Sumanth P. Reddy (ORCID: https://orcid.org/0000-0002-6275-6001)
- Diego Cuicapuza (ORCID: https://orcid.org/0000-0002-5735-4614)
- Brenda Ayzanoa (ORCID: https://orcid.org/0009-0009-2958-6563)
- Michael Wilson (ORCID: https://orcid.org/0000-0001-8983-0067)
- Jarod Morgenroth-Rebin
- Marlon P. Saavedra
- Oliver A. Bocanegra (ORCID: https://orcid.org/0000-0002-9938-6008)
- Katherine Lugo
- Antonio Chavez Neyra
Institutions
- University of California, San Francisco (US)
- Universidad Peruana Cayetano Heredia (PE)
- Hospital Nacional Dos de Mayo (PE)
- The University of Texas Medical Branch at Galveston (US)
Publication Details
- Journal
- medRxiv
- Published
- 2026-11-15
- DOI
- https://doi.org/10.64898/2026.11.13.26367457
- Primary Topic
- Bacterial Identification and Susceptibility Testing
- Type
- preprint