SPIN Score (Celiac Plexus Radiosurgery) NCT03323489
ForPancreatic cancer with retroperitoneal pain considered for celiac plexus SRS
TL;DRPost-hoc predictors of pain response after celiac SRS: response 32% / 53% / 89% by 0 / 1 / 2 SPIN points.
The actionable RT read is timing, not technique: neurotoxic chemo exposure carried OR 5.1 (p=0.009) against response, so the referral decision moves earlier in the disease course, before oxaliplatin or taxane neuropathy. Celiac SRS is NCCN-listed and single-fraction, so the gate is who you irradiate and when, not dose.
In pancreatic cancer pts with retroperitoneal pain scoring above 6 and no prior neurotoxic chemo, this supports considering celiac SRS earlier rather than after systemic lines; it does not tell you what to do for the chemo-exposed, low-pain pt, where response was 32%.
The gate on celiac SRS is patient selection and timing, not plan quality: neurotoxic chemo exposure carried OR 5.1 (p=0.009) against pain response, and 2-point pts responded 89% vs 32% at 0 points. Argues for taking the referral early rather than as last-line salvage.
The variable that mattered is one med onc controls: prior neurotoxic chemotherapy predicted failure of celiac SRS (OR 5.1, p=0.009). For a pt with severe retroperitoneal pain, this argues for a celiac SRS referral alongside, not after, an oxaliplatin or taxane backbone.
| SPIN score | n | Pain response |
|---|---|---|
| 0 | 31 | 32% |
| 1 | 40 | 53% |
| 2 | 19 | 89% |
10 details
Post-hoc analysis of the pivotal single-arm phase 2 celiac plexus radiosurgery trial (NCT03323489), n=90 evaluable. Candidate baseline predictors screened by univariate then multivariate logistic regression, with only multivariate-significant variables carried into the score. Internal validation by bootstrap resampling, 500 iterations, for an optimism-corrected AUC.
Pain response per parent protocol: a ≥2-point reduction in average pain on the Brief Pain Inventory-Short Form, baseline to three weeks. Parent-trial response was 53% (95% CI 42-64%).
Only neurotoxic chemotherapy exposure (OR 5.1, p=0.009) and baseline pain intensity (OR 1.8, p=0.003) survived multivariate analysis; age and therapy line dropped out to collinearity. The resulting 2-point score separated response into 32% / 53% / 89%, with AUC 0.716 apparent, 0.714 optimism-corrected.
| Predictor | Univariate | Multivariate |
|---|---|---|
| Neurotoxic chemo exposure | OR 5.33 (2.13-13.4), p<0.001 | OR 5.1, p=0.009 |
| Baseline pain intensity | OR 1.73, p=0.003 | OR 1.8, p=0.003 |
| Age | OR 1.06, p=0.014 | lost significance |
| Therapy line | OR 0.65, p=0.04 | lost significance |
The score was derived and internally validated in the same 90 pts, so the bootstrap corrects optimism from resampling but not from the variable-selection step that preceded it. The 2-point stratum is n=19, and the dichotomy at pain >6 was picked from the same data that shows steeper response above 7 and 8.
The neurotoxic-chemo term is the interesting one because it is a timing variable a clinician controls, not a fixed patient trait: it implies referral for celiac SRS competes with the systemic sequence rather than following it. Whether that reflects true nerve injury blunting an ablative pain response, or confounding by later-line disease biology, the post-hoc design cannot separate.
Post-hoc derivation on the parent trial's own 90 pts; internal bootstrap only, no external cohort. Design dominates the read regardless of effect size.
- External validation of the SPIN score in an independent cohort
- Is neurotoxic chemo effect causal or a proxy for later-line disease
- Durability of pain response beyond the 3-week endpoint
📚 Sources · 🐦 1 tweet
Celiac SRS = convenient, effective tx for intractable pain, but who benefits most?
— Dr. Nina Niu Sanford (@NiuSanford) May 30, 2026
Post-hoc Ph2 analysis identified 2 response predictors (SPIN score): severe baseline pain & no prior neurotoxic chemo.
Supports earlier use before potential chemo neuropathy. #ASCO26 @OncoAlert pic.twitter.com/3qFYU6N1iD
The longer read
Celiac plexus radiosurgery entered NCCN on the strength of a single-arm phase 2 with a 53% pain response rate, and the honest problem with a 53% number is that it describes nobody: half the pts in front of you get meaningful relief and half undergo a procedure for nothing, with no way to tell them apart at consult. That is the gap this analysis is trying to close, and it is worth taking seriously as a question even while treating the specific answer as provisional.
What makes the result more than a routine subgroup exercise is which variable survived. Baseline pain intensity predicting response is close to a statistical inevitability: a pt scoring 9 has room to drop 2 points on the BPI and a pt scoring 4 has much less, so a fixed-threshold response definition mechanically favours the severe end. Floor effects of this kind appear in nearly every palliative pain trial and they say more about the endpoint than the intervention. Neurotoxic chemotherapy exposure is a different kind of finding. It is not a baseline trait of the patient, it is a consequence of the sequence the oncology team chose, and it carried the larger effect (OR 5.33 univariate, 5.1 on multivariate) while age and therapy line washed out. If the association is causal, celiac SRS is a treatment whose efficacy is being spent down by the systemic therapy that precedes it, which is a genuinely different framing from the usual palliative-RT question of dose and technique.
The skeptical reading is that neurotoxic exposure is a proxy for time. Pts who have seen oxaliplatin or a taxane are further into their disease course, with more established central sensitisation and more opioid tolerance, and the collinearity that knocked out therapy line is consistent with exactly that. Nothing in a post-hoc analysis of 90 pts can adjudicate between a nerve-injury mechanism and a later-line confounder, and the two carry different clinical implications: the first argues for early referral in everyone, the second only for recognising that late-line pts respond worse to most things.
The methodology should move confidence modestly and in a specific direction. The bootstrap is the right instinct and the optimism-corrected AUC of 0.714 sitting essentially on top of the apparent 0.716 looks reassuring, but that correction addresses resampling optimism in a fixed model, not the selection that produced the model, and the >6 pain cutpoint was chosen from the same 90 pts that show response climbing to 83.3% above 7 and 85.7% above 8. An AUC near 0.71 is modest discrimination in absolute terms, and the 89% figure that carries the abstract rests on 19 pts. Expect the extremes to regress in any external cohort, with the ordering probably holding and the spread narrowing.
The decision this moves is referral timing, not the RT plan itself. The prescription is unchanged; what changes is whether a pt with severe retroperitoneal pain is sent for celiac SRS before neurotoxic systemic therapy rather than after everything else has failed. That is a low-cost change to make on suggestive evidence, and it is the one place where acting ahead of external validation is defensible.