Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Data, Analytics & AI · EBM in Individualized Immuno-Oncology

Application of Evidence-Based Medicine in Individualized Immuno-Oncology

The FDA’s 2017 tissue-agnostic approval of pembrolizumab for microsatellite instability-high (MSI-high) or mismatch repair-deficient (dMMR) solid tumours was the first approval based on a biomarker rather than tumour location. It formalized a biomarker-driven model of evidence-based medicine in which treatment selection is individualized across cancer types rather than by organ of origin.

Pubrica AcademyWritten by an oncology evidence synthesis reviewer Reviewed byan immuno-oncology clinical reviewer Published10 October 2026 Last checked10 October 2026 Reading time7 minutes Sources10 peer-reviewed

Evidence-based medicine in immuno-oncology has shifted from evaluating treatments primarily by tumour type (lung cancer, melanoma, and colorectal cancer trials, each generating a separate evidence base) toward evaluating them by biomarker status, which may cross tumour types entirely. This shift was formalized by a specific, well-documented regulatory milestone: in 2017, pembrolizumab became the first cancer treatment that the FDA approved on the basis of a biomarker rather than the anatomical site in which the tumour originated, specifically for solid tumours with microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) status1. An understanding of how that approval came about, and of what biomarker-driven evidence requires that tumour-type-driven evidence does not, is central to the application of evidence-based medicine in individualized immuno-oncology.

Learning objectives

  • Features that made the 2017 pembrolizumab MSI-H/dMMR approval the first of its kind
  • The evidentiary standard required by the FDA for a biomarker-based, tissue-agnostic approval
  • Extension of the biomarker-driven model by tumour mutation burden in a 2020 approval

Common misconceptions

  • “Cancer treatment evidence is organized primarily by where in the body the tumour originates.” The FDA’s 2017 approval of pembrolizumab for MSI-H or dMMR solid tumours was granted regardless of tumour origin; it was the first time the agency approved a cancer treatment on the basis of a common biomarker across tumour types rather than tissue of origin. A comparable tumour-agnostic approval for TMB-high solid tumours followed in 2020, establishing biomarker status, rather than organ location, as the organizing evidentiary category for a growing share of immuno-oncology treatment decisions.

Formalization of biomarker-driven evidence by the 2017 approval

In May 2017, the FDA granted accelerated approval to pembrolizumab for adult and pediatric patients with unresectable or metastatic solid tumours identified as MSI-H or dMMR that had progressed following prior treatment and had no satisfactory alternative treatment options, or for MSI-H or dMMR colorectal cancer that had progressed after specific prior chemotherapy regimens1. This approval differed structurally from essentially every previous oncology drug approval in that it applied across tumour types defined by organ of origin (lung, colorectal, endometrial, gastric, and others were all represented), united only by a shared genomic biomarker rather than a shared anatomical site1. This was explicitly the first time the FDA had approved a cancer treatment on this biomarker-first basis, building directly on earlier clinical evidence that PD-1 blockade produced durable responses specifically in mismatch-repair-deficient tumors regardless of their site of origin7, and it represented a formal regulatory acknowledgment that, for tumours sharing this specific DNA mismatch-repair deficiency, the biomarker was a more clinically meaningful organizing category for treatment decisions than the tissue of origin, a classification that itself depends on standardized mismatch-repair and microsatellite-instability testing methodology4.

The biological rationale for this biomarker merits explicit statement, since it gives the tissue-agnostic approval its scientific coherence and distinguishes it from an arbitrary regulatory convenience. Mismatch repair deficiency impairs the capacity of a cell to correct errors introduced during DNA replication, causing mutations to accumulate at an unusually high rate; tumours with this deficiency consequently tend to carry a very high mutational burden and, as a direct result, produce an unusually large number of neoantigens, mutated proteins that the immune system can potentially recognize as foreign. The same biological logic connects tumour mutation burden to checkpoint-inhibitor response, and MSI-H/dMMR status and high TMB frequently, though not universally, co-occur in the same tumours. This partly explains why both biomarkers, despite measuring somewhat different characteristics, point toward the same class of treatment and were approved through a structurally similar tissue-agnostic regulatory pathway within a few years of each other, a pathway later extended to NTRK gene fusions, a third, structurally distinct biomarker given its own tissue-agnostic approval8.

Evidentiary standard required for a tissue-agnostic approval

Because an approval of this kind cannot rely on a single, tumour-type-specific randomized trial in the traditional sense, since by definition it spans multiple tumour types, each represented by a smaller patient subset, its evidentiary basis instead relies on pooling response data across multiple single-arm or basket trials9, in this case across 15 clinical trials, with the key supportive evidence being a consistently observed, clinically meaningful objective response rate specifically among the MSI-H/dMMR subgroup across those different tumour types1. This evidentiary model places a specific burden on diagnosis: because the approval depends entirely on the correct identification of MSI-H or dMMR status, the reliability and standardization of the testing methodology used to classify a tumour as biomarker-positive become as evidentially important as the treatment response data, a dependency that did not exist in the same way for traditional, tumour-type-specific approvals, in which the diagnosis (confirmation of the tumour’s organ of origin and histology) was comparatively more straightforward and more standardized across testing laboratories.

Extension of the model to a second biomarker in 2020

The same tissue-agnostic, biomarker-driven evidentiary model was extended in June 2020, when the FDA approved pembrolizumab for adult and pediatric patients with unresectable or metastatic tumour mutation burden-high (TMB-H, defined as ≥10 mutations per megabase measured by a specific companion diagnostic assay) solid tumours that had progressed following prior treatment and had no satisfactory alternative options2. As with the MSI-H/dMMR approval, this extended biomarker-driven individualization to a second, distinct genomic characteristic, and it similarly tied the defining threshold of the approval to a specific, named companion diagnostic rather than a universal, assay-independent cut-off, following the same approach used to manage the cross-platform measurement variability that complicates TMB assessment more broadly2, and the FDA maintains a public list of the diagnostic devices authorized as the companion test for each such approval10. In practice, the application of evidence-based medicine to individualized immuno-oncology now entails determining which biomarker-defined evidence base a patient’s tumour qualifies for (MSI-H/dMMR status, TMB-H status, PD-L1 expression level, or a tumour-type-specific trial result) against a structured actionability framework such as the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT), which ranks genomic alterations by the strength of the clinical evidence supporting a given biomarker-treatment pairing56, rather than assuming that a single evidence base organized solely by organ of origin covers the relevant treatment decision. The National Comprehensive Cancer Network (NCCN) guidelines reflect the same shift, incorporating biomarker-defined treatment pathways directly into tumour-specific guideline algorithms rather than treating biomarker testing as a separate, optional step3.

These two approvals also illustrate a broader principle concerning how evidence-based medicine handles a biomarker that is prevalent but not universal within any single tumour type, a situation quite different from a traditional approval in which essentially every patient with a given diagnosis is eligible for the studied treatment. MSI-H/dMMR status occurs in only a minority of patients within most of the tumour types in which it is found, commonly cited as ranging from a few percent in colorectal cancer to a somewhat higher proportion in endometrial cancer; a tumour-type-specific trial population would therefore be diluted by a majority of biomarker-negative patients unlikely to benefit from the same degree of immune-checkpoint response. Pooling biomarker-positive patients across multiple tumour types, as the pivotal evidence for the 2017 approval did, was accordingly not merely a convenient means of assembling a larger sample; it was the only practical means of generating a clinically meaningful evidence base for a patient subgroup too small within any single tumour type to support a dedicated, tumour-specific randomized trial. This is a structural feature of individualized, biomarker-driven oncology evidence that researchers and clinicians should recognize: the rarity of a clinically important biomarker within a given tumour type is often the reason its evidence base had to be built across tumour types, rather than a reason to discount the strength of support for the biomarker-specific finding.

Applying this evidence-based model in practice also requires acknowledging its current limits. Biomarker-driven, tissue-agnostic evidence exists for a specific and still limited set of biomarkers, with MSI-H/dMMR and TMB-H chief among the FDA-recognized examples, whereas PD-L1 expression level functions somewhat differently, as a tumour-type-specific rather than fully tissue-agnostic predictive marker in most current approvals. For the much larger number of other candidate biomarkers under active research, such as gene-expression signatures, specific neoantigen profiles, and circulating tumour DNA dynamics during treatment, the evidence base currently remains organized largely by tumour type, because the pooled, cross-tumour evidence needed to support a tissue-agnostic approval has not yet been generated or has not yet met the FDA’s evidentiary threshold. Clinicians and researchers applying evidence-based medicine to an individual patient’s immuno-oncology care should therefore determine, biomarker by biomarker, which evidentiary model currently applies, rather than assuming that, because one biomarker (MSI-H/dMMR) has reached tissue-agnostic status, the broader field of predictive biomarkers in immuno-oncology has made the same transition uniformly.

This process has direct practical implications for the structure of an evidence-based literature search for an individual patient’s case. Rather than searching only within the patient’s specific tumour type, which risks missing tissue-agnostic evidence generated primarily in other tumour types that nonetheless applies given a shared biomarker, a thorough search strategy should follow two parallel tracks: one organized by the patient’s specific tumour type and histology, covering the conventional, tumour-type-specific trial literature and guideline recommendations, and a second organized explicitly by any confirmed biomarker result (MSI-H/dMMR, TMB-H, or another validated marker), covering the cross-tumour, tissue-agnostic evidence base described above. Combining both tracks, rather than relying on the tumour-type search alone by default, allows a literature review to identify the full set of evidence-based options for which a biomarker-positive patient may be eligible.

Key points

Key points

Evidence-based medicine in immuno-oncology increasingly organizes treatment decisions around validated biomarkers, including MSI-H/dMMR and TMB-H, rather than around tumour origin alone. This shift was formalized by the FDA’s 2017 and 2020 tissue-agnostic approvals, both of which tie their evidentiary threshold to a specific, named companion diagnostic rather than to a universal, assay-independent cut-off.

Practical application

Practical application

  1. The applicable biomarker-defined evidence base should be identified before reliance is placed on a tumour-type-specific guideline alone. MSI-H/dMMR and TMB-H status can each independently qualify a tumour for a tissue-agnostic treatment option, regardless of organ of origin.
  2. The companion diagnostic assay and threshold to which a biomarker approval is tied should be verified. Both major tissue-agnostic approvals discussed here specify a named assay and cut-off rather than a universal, assay-independent standard; the use of a different, unvalidated assay may result in misclassification.
  3. Biomarker testing should be treated as integral to the evidentiary pathway rather than as a separate, additional step. Because tissue-agnostic approvals depend entirely on correct biomarker classification, the reliability of testing is as important to the evidence base as the treatment response data themselves.

Frequently asked questions

It was the first FDA cancer treatment approval granted based on a shared biomarker (MSI-H or dMMR status) across multiple tumour types, rather than being limited to tumours originating in one specific organ.

It relies on pooling response data from multiple trials or sub-groups across different tumour types that share the relevant biomarker, rather than a single trial population defined by tumour type, which places added evidentiary weight on the reliability of the biomarker test itself.

Both are genomic biomarkers used for tissue-agnostic approval, but they measure different characteristics (mismatch-repair function for MSI-H/dMMR and overall mutation count for TMB-H), and each approval is tied to its own specific companion diagnostic and threshold.

No; NCCN and similar guidelines incorporate biomarker-defined pathways alongside tumour-type-specific recommendations, so a biomarker-driven option is typically considered within, not instead of, the broader tumour-specific guideline framework.

Because cross-platform measurement variability, particularly for TMB, can produce different classifications of the same tumour depending on the assay used, tying an approval to a specific, validated diagnostic avoids applying a threshold against which the clinical trial evidence for the approval was never tested.

References

  1. U.S. Food and Drug Administration. FDA grants accelerated approval to pembrolizumab for first tissue/site agnostic indication. 23 May 2017. read 10 October 2026. View on FDA.gov
  2. U.S. Food and Drug Administration. FDA approves pembrolizumab for adults and children with TMB-H solid tumors. 17 June 2020. read 10 October 2026. View on FDA.gov
  3. National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology. read 10 October 2026. View on NCCN.org
  4. ASCO endorsement of College of American Pathologists guideline: Mismatch Repair and Microsatellite Instability Testing for Immune Checkpoint Inhibitor Therapy. read 10 October 2026. View on PMC
  5. European Society for Medical Oncology (ESMO). ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). read 10 October 2026. View on ESMO
  6. Mateo J, Chakravarty D, Dienstmann R, et al. A framework to rank genomic alterations as targets for cancer precision medicine: the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Ann Oncol. 2018;29(9):1895-1902. read 10 October 2026. View on PMC
  7. Le DT, Uram JN, Wang H, et al. PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. N Engl J Med. 2015;372(26):2509-2520. read 10 October 2026. View on PubMed
  8. U.S. Food and Drug Administration. FDA approves larotrectinib for solid tumors with NTRK gene fusions. read 10 October 2026. View on FDA.gov
  9. Basket Trials in Oncology: A Trade-Off Between Complexity and Efficiency. read 10 October 2026. View on PMC
  10. U.S. Food and Drug Administration. List of Cleared or Approved Companion Diagnostic Devices (In Vitro and Imaging Tools). read 10 October 2026. View on FDA.gov

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 10). Application of Evidence-Based Medicine in Individualized Immuno-Oncology. Pubrica. https://pubrica.com/academy/evidence-based-literature-review/evidence-based-medicine-in-individualized-immune-oncology/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Biomarker-Driven Treatment Pathway Checklist

This guidance applies to the evaluation of individualized immuno-oncology treatment options against the evidence base.

[ ] Checked MSI-H/dMMR status alongside tumour-type-specific evidence
[ ] Checked TMB-H status (and the specific assay/threshold used) alongside tumour-type-specific evidence
[ ] Verified the biomarker test used matches a validated companion diagnostic for the relevant approval
[ ] Checked current NCCN or equivalent guideline pathway for biomarker-defined options
[ ] Confirmed whether the biomarker-driven option is tissue-agnostic or still tumour-type-restricted for this specific indication

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s evidence-based literature review service maps a tumour’s biomarker profile against current tissue-agnostic approvals and guideline pathways, so that searches for treatment evidence reflect the current biomarker-driven landscape rather than tumour-type literature alone.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

Assessment of the biomarker-driven evidence base alongside tumour-type evidence

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.