Rising Risk for Leptomeningeal Metastases in NSCLC Patients with EGFR Mutations

Leptomeningeal metastases are more common in patients with non&ndash;small cell lung cancer who harbor <em>EGFR </em>mutations.&nbsp;Tyrosine kinase inhibitors were found to be the optimal treatment method for these patients, especially if they have not yet received a TKI treatment.

Leptomeningeal metastases (LM) are more common in patients with non—small cell lung cancer (NSCLC) who harborEGFRmutations, according to a Chinese study published in theJournal of Thoracic Oncology. Tyrosine kinase inhibitors (TKIs) were found to be the optimal treatment method for these patients, especially if they have not yet received a TKI treatment.

While LM occur in approximately 1% to 9.1% of all solid tumor patients, LM have increased in patients with NSCLC especially over the past 10 years. Yet there is no current standard-of-care approach to treating these patients. Current treatments include chemotherapy, TKIs, and WBRT.

The study screened a database of 5,387 lung cancer patients forEGFRgene status and incidence of LM to find an evaluable 160 treated patients. Of 3,775 patients who were tested forEGFRstatus, 1,258 patients hadEGFRmutations, including 576 with an exon 19 deletion (del 19) and 511 patients with the exon 21 Leu858Arg mutation (L858R).

Of all the screened patients, 184 (3.4%) had LM, and only 160 had a knownEGFRstatus, of which only 118 had anEGFRmutation and 109 patients had commonEGFRmutations, including 53 with del 19 and 56 with L858R.

The median age of the evaluable patients was 57 (range, 28-78), and 85.6% of the patients had stage IV NSCLC. Brain metastases were also found in 103 patients, 59 of which were diagnosed with brain metastases at the same time as the LM.

Patients had undergone therapy for NSCLC prior to the LM diagnosis, including 32 who had received TKI therapy alone, 6 who received chemotherapy, and 51 patients who received a combination therapy. Seven patients had WBRT for brain metastases prior to diagnosis of LM.

For treatment of LM, 49 patients (45%) received a TKI treatment alone, 6 (5.5%) underwent WBRT alone, and 42 patients received a combination therapy. Eighty-eight total patients received TKI treatment—gefitinib (36.4%), erlotinib (40.9%), icotinib (17%), or another TKI (5.7%)—either alone or as a combined therapy.

Patients with wild-typeEGFRwere more likely to undergo WBRT (16 of 42 patients, 38.1%), whereas 7 patients received chemotherapy, 6 received TKI treatment, and 13 received best supportive care.

The median overall survival (OS) for all evaluable patients after LM diagnosis was 8.7 months (95% CI, 7.3-10.1). Patients withEGFRmutations had a longer OS compared with patients with wild-typeEGFRof 8.9 months (95% CI, 7.2-10.7) versus 7.3 months (95% CI, 4.6-10), although it was not significantly different (P= .686).

Patients treated with a TKI for LM (n = 88) had a longer OS of 10 months (95% CI, 8.9-11.1) than those that did not (3.3 months; 95% CI, 0.5-6.1;P<.001). The longevity of the TKI treatment was increased if the patient had not been on a TKI prior to diagnosis of the metastases. Median OS for patients initiating TKI at LM diagnosis was 12.2 months (95% CI, 9.7-14.8) compared with 9.2 months (95% CI, 7.8-10.5) for those who had progressed on a prior TKI treatment (P= .016).

“This may be explained by the theory that a greater effect on activated tumor cells may be seen on the first exposure to TKIs because the acquired resistance clones did not predominate,” the authors noted in the study.

Forty-two total patients underwent WBRT and demonstrated a median OS of 9.3 months (95% CI,8.4-10.3) compared with OS of 8.1 months (95% CI, 4.8-11.4) for those who not did receive brain radiation. There was no significant difference in OS between the patients who received combination WBRT and TKI treatment and the patients who only received TKI therapy, 9.7 months versus 10.1 months, respectively (P= .778).

Patients receiving chemotherapy showed an OS of 21 months (95% CI, 14.8-27.1), although these results only applied to a small pool of 14 patients who received chemotherapy after LM diagnosis.

Multivariate analysis using a ratio Cox proportional hazard model showed that TKI treatment (HR = 0.218; 95% CI, 0.116-0.411;P<.001) and chemotherapy (HR = 0.206; 95% CI, 0.092-0.460;P<.001) after LM diagnosis were both associated with better OS rates.

The study authors noted that the inability of certain TKIs to adequately penetrate the cerebrospinal fluid was one possible reason for why the frequency of LM in NSCLC patients withEGFRmutations has increased, as well as the longer survival periods of NSCLC patients, but more research is needed to determine the true cause.

Further studies could also examine treatment according to which TKI was prescribed, as erlotinib has a higher CSF penetration rate than gefitinib, therefore erlotinib would potentially be more effective in the management of LM.

“This study had some limitations, however,” the study authors wrote in a statement. “We showed that OS after LM was longer than that in previous reports, and LM were much more frequent in NSCLC patients harboringEGFRmutations. EGFR-TKIs were the optimal strategy for LM withEGFRmutations, especially TKI treatment-naÏve patients. Nevertheless, active treatment with WBRT, with or without EGFR-TKIs, was not supported by our study.”


Li Y-S, Jiang B-Y, Yang J-J, et al. Leptomeningeal metastases in non-small cell lung cancer patients with EGFR mutations. [Published online August 14, 2016.]J Thorac Oncol.doi: 10.1016/j.jtho.2016.06.029.