|
Stereotactic
radiosurgery (SRS) with or without whole-brain radiation therapy (WBRT)
is an approved management option for both single and multiple brain
metastases. It provides
69–91% actuarial 1-year local tumor control rate, and median
survival after treatment varies in different series from 7 to 15
months. Recursive partitioning analysis (RPA) of various
clinical variables permitted definition of three prognostic classes
based on the patient age, Karnofsky performance scale (KPS) score,
control of primary cancer, and presence of distant metastases beyond
the brain. The prognosis is generally poor in RPA Class III, which
incorporates all patients with KPS scores <70, whereas those
individuals who are completely dependent on relatives or medical
personnel are frequently considered not suitable for radiosurgery
It should be
noted, however, that RPA Class III is a heterogeneous group, and
presence or absence of other prognostic variables besides KPS score,
such as patient age, ambulatory performance status, multiplicity of
brain metastases, response to steroids, control of primary cancer,
and presence and location of extracranial metastases may
significantly influence the treatment outcome. Moreover, poor
performance status in a patient with brain metastases may result
either from presence of intracranial tumor(s) or uncontrolled
somatic disease. The cause of low KPS score (cerebral vs.
extracerebral), however, was not previously taken into account for
determination of patient prognosis after irradiation of brain
metastases. The objective of this study was evaluation of SRS in
patients with metastatic brain tumors and KPS score ≤50 with a
particular emphasis on prognostic significance of the determined
cause of poor performance status.
Management
options of intracranial metastases in patients with low KPS score
are scarce. Many propose withholding active treatment in such cases,
particularly with an active systemic disease and provide only
symptomatic therapy with steroids or anticonvulsants. Steroids can
result in dramatic improvement of the general condition, and
response to dexamethasone may be predictive of survival, but
duration of such an effect is usually short, and significant
complications may accompany prolonged high-dose therapy. Whole-brain
radiation therapy, which is considered as a standard option for
management of intracranial metastases not amenable to surgery or
radiosurgery, can provide only limited prolongation of survival of
RPA Class III patients. In approximately two thirds of cases with
serious neurologic dysfunction, clinical improvement within 1–2
weeks can be expected, but the treatment may be associated with
significant toxicity, which may result in further patient
impairment. Chemotherapy can be effective in some patients with
brain metastases; however, its survival benefit has not been
demonstrated to date, and on the basis of current evidence, this
treatment is not considered for the RPA Class III cohort. Finally,
good performance status is one of the main selection criteria of the
surgical candidates. In general, microsurgical tumor excision is
indicated for large accessible metastatic brain tumors with
significant mass effect and progressive neurologic symptoms, but it
is not considered beneficial if life expectancy is <2–3 months. The
addition of surgery to WBRT may not improve survival of patients
with brain metastases and active extracranial disease, although it
may favorably influence quality of life. Median survival after
surgical removal of metastatic brain tumors in the RPA Class III
cohort varies from 1.4 to 5 months, whereas mean predicted risk of
major complications after surgery for metastatic brain tumor in
individuals aged ≥65 years with preoperative KPS score of 50 and
functional Grade III neoplasm is 23%.
Stereotactic
radiosurgery offers the potential of low-risk treatment for patients
with comorbidities that preclude open surgery and for those who have
surgically inaccessible lesions. It seems that survival rates after
radiosurgical and surgical management of brain metastases may be
comparable. Compared with WBRT, the magnitude of survival gain after
radiosurgery might be the greatest in patients with poor prognostic
characteristics. Sanghavi performed a retrospective analysis
to estimate the survival difference between patients with newly
diagnosed brain metastases who underwent SRS in addition to WBRT and
those who were treated with WBRT alone. The pooled data from the
databases of 10 radiosurgical centers were compared with the
Radiation Therapy Oncology Group results in each RPA class.
Addition of SRS to WBRT
resulted in a statistically significant increase of the median
survival from 7.1 to 16.1 months in the RPA Class I cohort, from 4.2
to 10.3 months in the RPA Class II cohort, and from 2.3 to 8.7
months in the RPA Class III cohort. Addition of local
treatment, either surgery or radiosurgery, to WBRT increases the
probability of neurologic improvement and its duration. Radiosurgery
can be easily done as a 1-day treatment on an outpatient basis,
which offers more opportunities for active treatment of systemic
disease and avoids protracted fractionated radiotherapy, which may
be important in patients with limited life expectancy. In addition,
radiosurgery seems to be more cost-effective than open surgery for
brain metastases, which may be especially evident in patients with
low KPS score who are at high risk of postoperative complications.
Nevertheless, SRS is usually considered not suitable for patients
with metastatic brain tumors who are in poor general condition. The
proportion of RPA Class III patients in the reported radiosurgical
series varies from 5–20%, and few of these include individuals with
KPS scores <50.
Weltman
investigated results of radiosurgical management of metastatic brain
tumors in cases with SIR of 0–5, which is more or less comparable
with our cohort. Among 44 analyzed patients, none had KPS scores
<50. The median survival in their series was 4.5 months, the
probability of 6-month survival was 35.6% ± 7.1%; longest survival
was 14 months. The authors postulated that “patients surviving less
than 6 months after [radiosurgical] treatment might not have enough
time to enjoy its potential benefits” and concluded that “although
[radiosurgery] may offer some improvement in quality of life, and
perhaps in survival time, for patients with a very poor prognosis,
[it] may not justify the stress, cost, and risk involved.” Such a
statement, however, seems arbitrary to us.
Results of our
study demonstrate that SRS may be considered a reasonable option for
some patients with low KPS scores and without urgent surgical
indications. Mean and median survival after treatment were 3.8 and
3.3 months, respectively, which is comparable to previously
published outcomes in RPA Class III cohorts. The probability of
6-month survival was 0.20 ± 0.05, and the longest survival was 11.5
months. No significant complications or side effects were noted.
Moreover, referring physician or patients’ family members noted some
clinical improvement in 60% of cases during follow-up, although
objective evaluation is not possible because of the retrospective
nature of this study. It should be noted that 62% of patients in our
series had tumors located in near-eloquent or eloquent brain areas,
which made them poor surgical candidates, whereas in 33% of cases, SRS was done as salvage treatment.
Neither functional grade of the tumor nor previous management of
intracranial metastases showed statistically significant
associations with outcome. Meanwhile, two important prognostic
variables were identified by multivariate statistical analysis: the
presence of distant metastases beyond the brain and the cause of the
poor performance status of the patient. In fact, extracranial
disease status, defined as a control of the primary cancer, presence
of extracranial metastases, or both, consistently showed
statistically significant associations with the outcome after radiosurgical or radiotherapeutic management of brain metastases ,
and therefore it was included in the majority of existing prognostic
schemes. At the same time, despite the fact that poor performance
status in a patient with brain metastases may result from either
presence of intracranial tumor(s) or uncontrolled somatic disease,
the cause of low KPS score (cerebral vs. extracerebral) has not
previously been taken into account for determination of the patient
prognosis. |
