In a culmination of over 10 years of research effort, dating back to original observations that the hippocampus in rodent pups is an exquisitely radiosensitive structure, and central to the generation of memory-forming neurons from an innate radiosensitive stem cell compartment, Gondi et al have produced level I evidence of its seminal role in humans. In the plenary session at the SNO 2018 Annual Meeting (Society for Neuro-Oncology 23rd Annual Meeting. New Orleans, LA; Nov 15–18, 2018), they presented preliminary data from a phase III randomized trial with 500 plus patients, NRG CC001, supporting this concept.1 Patients with brain metastases were randomized to whole brain radiotherapy with or without hippocampal avoidance, and dosimetric compliance was monitored through rigorous central review. Patients on both arms also received memantine, based on results from a prior RTOG trial.
Between July 2016 and March 2018, the trial enrolled 518 adult patients with brain metastases who were stratified by RPA class and receipt of prior radiosurgery/surgery. The whole brain radiotherapy dose was 30 Gy in 10 fractions of 3 Gy each. The hippocampal maximum dose was to be kept below 15 Gy (and ideally <13.5 Gy), and the dose to 100% of the hippocampus was to be kept below 8.5 Gy (and ideally <7.5 Gy). Standardized neurocognitive function (NCF) tests were performed at baseline, 2, 4, 6, and 12 months. The primary endpoint was time to NCF failure, defined as decline on at least one of the following tests using the Reliable Change Index: Hopkins Verbal Learning Test-Revised, Trail Making Test, or Controlled Oral Word Association. Cumulative incidence was used to estimate time to NCF failure (death without NCF failure was treated as a competing risk) with between-arms differences tested using Gray’s test.
As expected, the trial primarily enrolled older patients with non–small cell lung cancer (almost 60%; fewer than 20% had breast cancer) with multiple brain metastases, with the vast majority (>85%) falling into RTOG RPA prognostic class II. The treatment arms did not differ in baseline clinical characteristics, baseline NCF, overall high-grade toxicity outcomes, overall survival, or intracranial progression. The time to NCF failure, the primary trial endpoint, was significantly longer in favor of hippocampal-avoidant whole brain radiotherapy with memantine (P=.012). The 6-month NCF failure rates were 69.1% (95% CI, 61.8–75.3) vs 58.0% (95% CI, 50.2–64.9) for whole brain radiotherapy with memantine vs hippocampal-avoidant whole brain radiotherapy with memantine, respectively. After adjusting for stratification variables, hippocampal-avoidant whole brain radiotherapy with memantine (HR, 0.73; 95%CI, 0.56–0.94; P=.016) and age ≤61 years (HR, 0.61; 95%CI, 0.46–0.81; P=.0006) remained robustly significant. Consequential to the cognitive protection engendered by hippocampal-avoidant whole brain radiotherapy with memantine, patient-reported symptom burden was also lower in this arm. The trial accrued at a rate of 16 patients per month and reached the accrual goal 2 years ahead of planned completion.
Several salient observations emerge from this trial:
- Although there has been a significant increase in the utilization of stereotactic radiosurgery over whole brain radiotherapy over the last 3+ years because of the neurocognitive decline associated with whole brain radiotherapy, physicians and patients remain cognizant of the very high likelihood of intracranial relapse consequential to the use of radiosurgery alone, and therefore, NCF-preservation whole brain radiotherapy techniques, as employed in this trial, were met with enthusiastic support, completing enrollment 2 years ahead of schedule, a rather unusual circumstance in oncology trials.
- The trial is an elegant demonstration of diligent bedside-to-bench-to-bedside research. The famous 1957 case of HM, who underwent bitemporal medial lobectomies for intractable seizures resulting in severe anterograde amnesia, provided the first genuine clinical clues about the role of the hippocampus in memory functions. Subsequent rodent work by several researchers, including, but not limited to that by Monje et al, established the presence of an exquisitely radiosensitive stem cell compartment in the peri-hippocampal location, which subserves a critical role in neurogenesis for memory formation. Prior work by Tome et al established the clinical significance, and dosimetric parameters for hippocampal sensitivity in humans, and NRG Oncology, in a series of trials, demonstrated that both chemical modulation (using memantine), and dosimetric preservation, using hippocampal-avoidant techniques, reduces rates of cognitive decline in patients receiving cranial radiotherapy.
- Therefore, hippocampal avoidance for cranial malignancies well beyond brain metastases should now be considered whenever feasible. The contention here is logical; if the hippocampus of a patient with brain metastasis is sensitive to radiation, then the hippocampus of a patient with a low-grade glioma or another malignancy would also be radiation-sensitive; the underlying disease is not an excuse to not attempt hippocampal sparing. This now provides level I evidence for hippocampal sparing in patients with brain malignancies, thereby providing a fillip to technologies that can achieve this robustly, such as intensity-modulated photon and proton techniques.
- Finally, the hazard ratio for cognitive preservation from hippocampal avoidance (in the presence of memantine) in this trial is 0.74; in the prior NRG Oncology trial RTOG 0614, the hazard ratio for cognitive preservation from the addition of memantine in patients receiving whole brain radiotherapy was 0.78. Therefore, the overall hazard ratio for hippocampal avoidance with memantine, in the context of patients receiving whole brain radiotherapy, is the product of the two hazard ratios, or 0.58, compared with whole brain radiotherapy. This is a very robust improvement, similar to, or even exceeding, that achieved in trials comparing radiosurgery with whole brain radiotherapy, thereby reigniting interest in trials of radiosurgery alone versus radiosurgery plus hippocampal-avoidant whole brain radiotherapy plus memantine, where one would expect similar cognitive outcomes but superiority in terms of intracranial disease control when whole brain radiotherapy is utilized. Such trials are already on the drawing board.