Microbiome-derived TOPOSCORE is associated with response to immune checkpoint inhibitor re-induction after chemotherapy in primary resistant metastatic melanoma.
Haferkamp S, Schilling B, et al. • Cancer immunology, immunotherapy : CII • 2026
In this exploratory analysis of 24 PROMIT patients, baseline gut microbiome TOPOSCORE classification was insufficient to predict outcome of ICI re-induction after dacarbazine, longitudinal patterns were heterogeneous, and the small number of complete longitudinal profiles precludes conclusions that chemotherapy induced a favorable microbiome transition.
Key Findings
Results
Baseline TOPOSCORE distributions and s-scores were not associated with clinical outcome in patients receiving ICI re-induction after dacarbazine.
Exploratory analysis of 24 patients with longitudinal stool samples from the phase II PROMIT trial (NCT04225390).
Patients with primary ICI resistance received two cycles of dacarbazine followed by ICI re-induction.
Stool samples underwent shotgun metagenomic sequencing; ecosystem states were classified as SIG1- or SIG2-associated using the TOPOSCORE framework.
The authors conclude that baseline TOPOSCORE classification was insufficient to predict treatment outcome in this cohort.
Results
Among nine patients with complete baseline–chemotherapy–ICI (BL-CTX-ICI) sampling, most kept the same TOPOSCORE category while a minority changed category.
Six of nine patients retained the same TOPOSCORE category throughout treatment.
Three of nine patients changed category.
Longitudinal TOPOSCORE patterns were described as heterogeneous.
Results
The two patients with partial response (PR) showed different TOPOSCORE trajectories.
One PR patient remained SIG2 throughout treatment.
The other PR patient changed from SIG2+ to SIG1+ after chemotherapy and returned to SIG2+ during ICI re-induction.
Results
Analyses using all available samples at each treatment phase did not show significant differences in s-score among PR, SD, and PD groups.
This analysis was not restricted to patients with complete longitudinal sampling.
Comparisons were made across partial response, stable disease, and progressive disease groups.
Conclusions
The small number of complete longitudinal profiles precludes concluding that chemotherapy induced a favorable microbiome transition.
Only nine of 24 patients had complete BL-CTX-ICI sampling.
Authors state the findings are exploratory and support further evaluation of microbiome ecosystem dynamics in larger prospective cohorts.
What This Means
Some people with advanced melanoma do not respond to immune checkpoint inhibitors from the start. An earlier trial, PROMIT, found that giving a short course of chemotherapy (dacarbazine) and then restarting immunotherapy could restore benefit in some of these patients. This study asked whether the makeup of the gut microbiome, assessed using a scoring framework called TOPOSCORE that sorts gut bacterial communities into two ecosystem types (SIG1 or SIG2), was linked to who responded.
In 24 patients with stool samples, the TOPOSCORE measured before treatment did not distinguish patients who responded from those whose disease stayed stable or progressed. Among the nine patients with samples at all key time points, six stayed in the same category and three switched. The two patients with partial responses followed different paths: one stayed in the same category throughout, while the other switched after chemotherapy and then switched back during immunotherapy.
This research suggests that, in this small group, the baseline TOPOSCORE could not predict response to re-started immunotherapy after chemotherapy. The sample was too small to say whether chemotherapy shifts the microbiome in a helpful way. The authors present these as exploratory results that need testing in larger prospective studies.
Haferkamp S, Schilling B, Baumgart L, Gesierich A, Glutsch V, Drexler K, et al.. (2026). Microbiome-derived TOPOSCORE is associated with response to immune checkpoint inhibitor re-induction after chemotherapy in primary resistant metastatic melanoma.. Cancer immunology, immunotherapy : CII. https://doi.org/10.1007/s00262-026-04618-2