Meta-analysis topic ideas in oncology

Oncology produces new trials faster than reviews can keep up with, so a review that was complete two years ago can already be missing the studies that matter. The risk runs the other way too: a drug that is popular in the news may already have several competing meta-analyses.

Good oncology topics usually come from narrowing the question, by tumour type, line of treatment, biomarker or population, rather than from the broadest version of a popular question.

Find a oncology topic that has not been done

SynthGap searches the published literature, checks each candidate against existing systematic reviews, and proposes topics with the primary studies behind them. The free plan includes 3 discovery runs a month.

Start free

Where gaps tend to hide in oncology

Biomarker-defined subgroups

Trials often report results by marker status (for example PD-L1 level or a specific mutation) in different ways, and a pooled estimate by subgroup may not exist.

Toxicity and adverse events

Reviews focus on survival. Pooling grade 3 or higher adverse events, immune-related events or treatment discontinuation for the same trials is a different and often missing question.

Quality of life and patient-reported outcomes

These are measured with different instruments and are often a secondary endpoint, so they get less attention than survival.

Older and frail patients

Trials under-enrol them. Pooling the trials that did, or comparing treatment de-escalation approaches, can be a distinct topic.

Shorter or lower-intensity treatment

De-escalation and shorter-duration trials are non-inferiority designs. They are newer and often not yet pooled.

Supportive care

Exercise, nutrition, anti-nausea treatment and cancer-related fatigue interventions are tested in many small trials.

Outcomes and effect measures that pool well in oncology

Overall survival and progression-free survival: pool hazard ratios with generic inverse variance. If only Kaplan-Meier curves are published, methods exist to reconstruct approximate patient-level data from the curves.
Objective response rate: risk ratio or odds ratio. Check that the trials used the same response criteria (for example RECIST version).
Grade 3 or higher adverse events: risk ratios. Report the definition and grading system used.
Quality of life scores: standardised mean difference, with attention to the minimal important difference for the instrument.

Common traps in oncology reviews

Crossover to the experimental treatment can dilute the overall survival difference. Note how each trial handled it.
Progression-free survival can be assessed by investigators or by blinded independent review, and the two can differ. It is also not a validated surrogate for overall survival in every setting.
Immunotherapy curves often separate late or cross. A single hazard ratio can then be misleading, and restricted mean survival time may describe the effect better.
Immature overall survival data in early reports: check for updated follow-up before pooling.

Check before you commit

Before committing months to a topic, check that nobody has beaten you to it. Search PubMed with its systematic review filter, the Cochrane Library, the PROSPERO register (for reviews that are registered but not yet published) and Epistemonikos. Look at how recent the latest review is and whether new trials have appeared since its search date. Step-by-step guide.

Starting points to try in SynthGap

These are areas to explore, not claims that they are open. Some may already be well covered, and finding that out quickly is the point.

Treatment-related adverse events with immune checkpoint inhibitors in older adults
Exercise interventions for cancer-related fatigue during chemotherapy
Shorter versus standard duration of adjuvant treatment

Questions

How do I find an oncology topic that is not already covered?

Narrow the question: one tumour type, one line of treatment, one biomarker group or one outcome such as toxicity or quality of life. Then check recent systematic reviews on that exact combination, and the PROSPERO register for ones in progress.

Can I pool progression-free survival from different trials?

Yes, using hazard ratios, but only if the trials define progression the same way and assess it comparably. State whether assessment was by investigators or blinded independent review.

Is a network meta-analysis better than a pairwise one in oncology?

It can be, when several treatments have never been compared directly, but it relies on the trials being similar enough to compare. Many oncology questions are still best answered by a focused pairwise review.

Other specialties

This page is general methodological guidance, not medical advice, and it does not list specific unreviewed topics. SynthGap's suggestions are AI-generated from the published literature, so verify them before registering a protocol.