• Discussion

    In the present series of NFA patients, excellent local control (95% at 10 years) was achieved when immediate postoperative radiotherapy was applied in case of residual tumor. In comparison, local control was only 49% at 5 years and 22% at 10 years when a wait-and-see policy was followed. Importantly, immediate postoperative radiotherapy did not result in an additional need for conventional hormonal substitution treatment, or in an excess of epilepsy, cerebrovascular disease, and intracerebral malignancy in comparison to an expectant strategy. Furthermore, it is noteworthy that life expectancy was similar in both groups, and did not differ from the general Dutch population. Our survey thus suggests that immediate postoperative radiotherapy in case of residual NFA can be applied safely.
    Local control rate after immediate postoperative radiotherapy reported here agrees with other studies, showing that 82% to 97% of patients remained free of tumor regrowth after 10 years of follow-up5,6,20-22. Comparable with our data, a local control rate of only 40% to 70% at 5 years and of 15% to 50% at 10 years was documented previously when a wait-and-see policy was followed23-26. Importantly, despite protocolized follow-up with serial MRIs, a symptomatic recurrence was observed in 4 of 34 prospectively followed patients after a period of only 28 months27. In agreement, symptomatic recurrences were recently reported to be present in 6% to 21% of patients28. In the present study, salvage radiotherapy in case of regrowth was deemed clinically necessary after a median interval of 38 months after first surgery in 50% (14/28) of NFA patients; in 7 patients after a second and in 1 patient even after a third operation. A wait-and-see policy can be expected to result in a higher frequency of MRI and an increased frequency of re-operations, which likely results in emotional and social dysfunction29 as well as in additional health care costs. One should, therefore, be aware of the possible risks of an expectant policy in case of residual postoperative NFA.
    A frequently used argument to postpone postoperative radiotherapy is the possible development of radiation-induced hypopituitarism7. This supposition is mainly based on the results from a small series of 35 patients7. In that report, 50% of patients had already pituitary hormonal deficiencies before radiotherapy, which increased to 75% after this treatment.
    Patient characteristics at diagnosis and directly after surgery were similar in subjects with residual NFA who did and did not receive immediate postoperative radio-therapy in this series. It is of relevance, therefore, that our study clearly demonstrates that there was no difference in the need for thyroid hormone, glucocorticoids, sex-steroids, and vasopressin between the immediate postoperative radiotherapy group and the wait-and-see group with salvage radiotherapy. This lack of negative impact of immediate postoperative radiotherapy on pituitary function could not be attributed to bias caused by differences in hormonal deficiencies before and shortly after surgery, or in clinical characteristics between the groups. A potential shortcoming of our study is that we did not evaluate the frequency of growth hormone replacement therapy in each group. Such an analysis was not done because this treatment was introduced relatively late in the time frame of our evaluation period. Moreover, it is very likely that many patients in each group already had growth hormone deficiency shortly after surgery, given the high frequency of other hormonal deficiencies18,30.
    Radiotherapy could result in other unwanted side effects. The possible negative effect of radiotherapy on the development of cerebrovascular disease is frequently mentioned but still debated9,10,12,31. In the present series, the risk for cerebrovascular di-sease was not different between groups. The induction of intracranial malignancies and menin-giomas by radiotherapy is also debated14,15. In our cohort, no intracranial malignancies and one meningioma was diagnosed in a total of 90 irradiated patients.
    Another possible late side effect of radiation therapy is radiation optic neuropathy, but we have already documented that this is a very rare complication, provided fractionated radiotherapy is applied with a recommended total dose not exceeding 45 Gray in NFA patients8.
    In the present study, we did not evaluate the effect of radiotherapy on cognitive function and on quality of life. Previous studies have shown diminished cognitive function and impaired quality of life in newly diagnosed patients with NFA compared to healthy subjects32,33. A cross-sectional study demonstrated reduced cognition and some impairment in quality of life in a mixed group of patients with non-functioning and hormone secreting pituitary tumors who were treated with surgery and radiotherapy compared to patients who were treated with surgery alone34. Such an effect was not found in another report35. Furthermore, the contribution of postoperative radiotherapy to a possible decline in mental performance and quality of life is not well understood, because prospective data, which take the effects of both conventional pituitary hormone substitution and growth hormone replacement into account, are currently not available. Moreover, it can be expected that improvement in radiation treatment techniques will result in significantly lower radiation doses to the cerebral parenchyma36, with an assumed sparing effect on cognitive function.
    Several studies have addressed the question whether there is increased mortality in NFA patients, and to define the possible negative impact of radiotherapy on morta-lity in this patient category31,37-39. The interpretation of these data is difficult, because of inclusion of patients with hypopituitarism not due to NFA, the possible effect of (treated) deficiencies of conventional anterior pituitary hormones and anti-diuretic hormone, as well as the effect of growth hormone deficiency on mortality39. In the present series, log-rank analysis demonstrated that survival did not differ between patients, who received immediate post-operative radiotherapy and patients in whom a wait-and-see policy was followed. When all NFA patients were combined, life expectancy was similar to that observed in the general age- and sex-matched population from the Netherlands. In comparison, increased mortality has been observed in several31,37-39, but not in all surveys40 comprising pituitary patients due to various causes. The largest series available so far shows a modest excess in overall mortality in NFA patients, without a significant independent adverse impact of radiotherapy39. Several factors such as differences in time frame of patient surveillance, with follow-up being starting as early as 1946 to 1958 in some previous reports12 as well as the relatively low frequency of transcranial surgery and the lack of additional negative impact of radiotherapy on conventional pituitary hormone deficiencies in the present series, may explain part of the discrepancy.
    In conclusion, immediate postoperative radiotherapy in case of residual NFA provides a marked long lasting improvement of local control among patients with residual non-functioning pituitary adenoma compared to surgery alone, without an additional deleterious effect on pituitary function and life expectancy. Therefore, results of this study support to perform immediate postoperative radiotherapy in this patient category. The present results also underscore that immediate postoperative radiotherapy is not necessary in apparently complete resected NFA.

Comments are closed.