Background: In the US, invasive cervical cancer (CC) will affect 12,710 women/year with ~4290 deaths; global deaths from CC are ~275,000 women per yr. Though treatment options for recurrent/metastatic (met) CC have improved by adding bevacizumab to chemotherapy, pts with met CC will eventually have cancer progression (Tewari et al ,NEJM 2014), and effective therapies are few. Data from our own institution (Wright et al, Cancer 2013) found PIK3CA mutations in 32% of adenocarcinomas of the cervix (ACC) and in 35% of squamous cell cancers of the cervix (SCC). KRAS mutations were present in 17% of AC’s and 0% of SCC. Thus, therapies targeting the PIK3CA and RAS-ERK pathways have rationale for treating met CC; this phase II study is testing a MEK and an AKT inhibitor combination. Pre-clinical studies support dual PI3Ki and MEKi given pathway redundancy and negative feedback loops. Methods: Trametinib (GSK1120212) is a reversible/selective inhibitor of MEK1/MEK2. GSK2141795 is an AKT1-3 inhibitor. Objectives of this study: RECIST 1.1 activity of trametinib and GSK2141795 in pts with recurrent CC; other objectives include assess PFS and OS duration, toxicities, mutation and co-mutation rates of PI3K and RAS-ERK signaling pathway genes using high throughput mutational analysis on FFPE samples, and mutational status with clinical benefit from trametinib and GSK2141795. Eligibility: recurrent or met CC (any histology eligible), receipt of 1 prior chemotherapy, and up to 1 add’l regimen, no prior use of PI3K or RAS-ERK pathway inhibitor, ECOG PS 0-2, normal organ function, and availability of FFPE tissue. The treatment regimen is as follows: trametinib 1.5 mg and GSK2141795 50 mg, both given PO and daily. Cycle length is 28 days and pts are assessed for RR every 2 cycles. The study opened in 10/2013, and as of 2/2015, 12 pts have been enrolled. Target accrual is 35 pts; there is a 91% power to detect an improvement in RR from 0.07 to 0.22 using a one-sided 0.09-level exact binomial test. The null hypothesis will be rejected if 5 or more pts respond to trametinib and GSK2141795; the null hypothesis of 0.07 is based on a weighted average of the observed RR from the 11 cohorts enrolled to GOG 127 and 227 series. Clinical trial information: NCT01958112