Seizure outcomes and survival in adult low-grade glioma over 11 years: living longer and better

We enjoyed reading the letter by J. Pallud regarding our paper published in Neuro-Oncology Practice, and thank him for his kind words.

We agree that adult diffuse low-grade glioma (LGG) is a progressive disease without a specific treatment and we did say that these tumors “have an inherent tendency to transform to higher grades over time.” The study was carried out to determine whether there had been an improvement in survival and seizure control over an 11-year period in a single center, and the idea evolved from the emerging data in the literature and from our own “natural history series” quoted by Dr Pallud.¹ Our clinical experience showed that, although most patients with LGG remained well for many years without any specific interventions, all tumors steadily grew, and the vast majority of patients eventually died of tumor progression. Indeed, based on these and other data that emerged in studies conducted in the decade 2000 to 2010, it became clear that the philosophy of watchful waiting was never going to improve survival, even before the papers of Jakola et al^2,3 showing prolonged survival in patients receiving early resections. However, a recent Cochrane review looking at whether biopsy or resection is the best treatment strategy for LGG “was unable to determine this as no studies were included and only low-quality evidence from non-RCTs is available.” ⁴

With regards to the timing of surgery and survival, we defined early surgery as occurring within 1 year and when we pooled the data, as requested, we found a significant benefit for progression-free survival (P = .014) and a nonsignificant trend for overall survival (P = .19). With regards to correlating extent of resection with seizure outcomes, the original audit recorded only a total or partial resection. We agree with Dr Pallud’s suggestion that quantifying extent of resection on MRI would further strengthen our paper, but this is not possible because the MRI scans from the 2006 cohort are no longer in existence. Therefore, we would only be able to analyze the 2017 cohort scans and, because there were only 44 patients who had surgery, this is too small a number to carry out a meaningful comparison.

When pooling the results in this way, there is no guarantee that key demographics such as age or tumor location/laterality would be matched, which would confound our results and make them effectively uninterpretable. In contrast, it was possible to do a meaningful comparison between the cohorts at 2 distinct time points because the patients were so well matched. Similarly, the unmatched group sizes between those who had early surgery and those who did not in pooled analysis reduces the statistical power of the significance testing.

It will be interesting to repeat this study in about 10 years, now that adjuvant radiotherapy and chemotherapy are more routinely given following publication of the RTOG-9802 study in 2016,⁵ so that we can assess how evolving oncological treatment is affecting survival and seizure outcomes.

Finally, we concur with Dr Pallud that more effort is needed to improve the outcome of these patients. However, it would not be meaningful to compare our results with those reported by the French Glioma network.⁶ First, our seizure outcomes were not routinely measured 6 months after surgery but were a snapshot of patient status at the time of data collection; second, we required 1 year of seizure freedom to classify patients as Engel Class I; and third, our results were an average of those patients who had early resection and those who did not. However, as only 22% of our patients were seizure free in the 2006 cohort compared to 57% in the 2017 cohort, it is encouraging to see that the outcomes for these patients are improving, albeit there is still much more to be done.