Metformin and Carbohydrate Restriction With Platinum Based Chemotherapy In Stage IIIB/IV Non-Squamous Non-Small Cell Lung Cancer (NS-NSCLC)

Lung cancer is the leading cause of cancer related mortality in both men and women. In the
U.S. alone, an estimated 160,340 lung cancer related deaths occurred in 2012, accounting for
about 28% of all cancer related deaths. Approximately 85% of lung cancer is classified as
non-small-cell lung cancer (NSCLC) with roughly two-thirds of these patients presenting with
advanced disease. Histologically, NSCLC can be subdivided into adenocarcinoma, squamous cell,
large cell, and non-small cell lung cancer that cannot be further classified. Those tumors
that are not squamous (adenocarcinoma, large cell, not classified) are collectively termed as
non-squamous, non-small cell lung cancer (NS-NSCLC) and account for roughly 75% of all non
small cell cancer cases.

The most accepted upfront treatment for patients with advanced stage NSCLC has been platinum
based chemotherapy. Current standard practice for treatment of stage IV non-squamous NSCLC
patients with cytotoxic chemotherapy has evolved over the past decade. In a sentinel study in
2005, Sandler and colleagues demonstrated that the addition of bevacizumab to platinum
doublet chemotherapy (carboplatin/paclitaxel) followed by maintenance bevacizumab conferred a
survival advantage when compared to platinum doublet chemotherapy alone (12.1 mos vs. 10.
mos) in patients with stage IV non squamous, non-small cell lung cancer. Following this, the
largest phase III study ever conducted in stage IV NSCLC randomized more than 1700 patients
with stage IV lung cancer to either cisplatin/pemetrexed or cisplatin/gemcitabine. This study
was the first to reveal an interaction between chemotherapy and histology,demonstrating a
survival advantage for the subset of patients with non-squamous cell treated with
cisplatin/pemetrexed when compared to those treated with cisplatin/gemcitabine (11.0 vs 10
mos, p<0.05. Building upon this, a recent study evaluating maintenance pemetrexed (continuing
treatment after the four cycles of platinum doublet therapy) in non-squamous cell lung cancer
demonstrated a significant survival advantage for patients receiving maintenance pemetrexed
vs. placebo after four cycles of cisplatin/pemetrexed (13.9 mos vs. 11.0 mos, p<0.05). Based
on these studies, a regimen of cisplatin or carboplatin with pemetrexed followed by
maintenance single agent pemetrexed has become one of the most accepted frontline treatments
for patients with stage IV non-squamous, non small cell carcinoma.

Recently, there has been a firmer understanding of the relevant signaling pathways critical
for lung cancer growth, leading to the development of novel, targeted therapies. The
discovery of the EGFR and ALK pathways in lung cancer and the subsequent development of drugs
that target these pathways, erlotinib and crizotinib respectively, has yielded unprecedented
survival times in stage IV non-squamous, non small cell lung cancer. Unfortunately, only 25
to 30% of patients harbor these mutations, and platinum based chemotherapy remains the
cornerstone of treatment for the 60-70% of patients with stage IV disease without
identifiable targets. In attempts to improve outcome, a large need remains to develop novel,
effective agents to combine with platinum therapy that possess a favorable toxicity profile
at a reasonable cost.

Metformin:

Metformin, an oral biguanide agent used for the treatment of non-insulin-dependent diabetes
mellitus, is now prescribed to more than 120 million people worldwide. Its glucose lowering
effects result from both inhibition of liver gluconeogenesis and increased insulin
sensitivity in peripheral tissue. Metformin has limited adverse effects with little or no
risk of hypoglycemia in healthy, nondiabetic controls. In addition to its anti-diabetic
properties, metformin has demonstrated both chemopreventative and therapeutic effects in both
prostate and breast cancer. Jiralersprong et al reported that diabetic patients with breast
cancer receiving metformin had a 24% complete pathological response rate to neoadjuvant
chemotherapy compared to only 8% of those in the non-metformin group. More recently Joshua et
al reported reduced tumor Ki-67 rate as well as significant reductions in fasting glucose,
insulin growth factor 1 and BMI (body mass index) in prostate cancer patients receiving
neoadjuvant metformin prior to radical prostatectomy. Large epidemiological studies
consistently have shown substantially lower incidence of cancer occurrence and death in
diabetic patients taking metformin compared to those receiving other therapies. Most
recently, a retrospective study of patients with ovarian cancer found that 5-year
disease-specific survival was significantly better for diabetic patients who took metformin
than for those who did not (67% vs 47%; P = .007). Based on these important observational
studies, there are currently several, ongoing prospective studies evaluating metformin in
nondiabetic patients with both early stage and late stage breast cancer and prostate cancer,
respectively.

The role of metformin as a preventative and therapeutic agent in lung cancer is beginning to
be assessed. A recent epidemiological study from Taiwan demonstrated a 39-45% decreased risk
of lung cancer in diabetic patients being treated with antidiabetic drugs including metformin
versus those not taking these agents. These studies have triggered preclinical and clinical
observational trials that further support metformin's potential as an antineoplastic agent.
In a recent in vitro study, Ashinuma et al. was able to demonstrate inhibited clonogenicity,
cell growth and proliferation in four different human lung cancer cell lines exposed to
variable concentrations of metformin. These and other preclinical data have triggered in vivo
experiments in mice. Memmott and colleagues showed that oral administration of 1 or 5mg/mL of
metformin decreased lung tumor burden by 38% and 53% respectively in A/J mice injected with
tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1 byutanone (NNK). Importantly, steady
state levels of metformin were similar to those achieved in diabetic patients using
metformin, suggesting the treatment and prevention of lung cancer could be achieved with
standard oral dosing. Finally, two observational studies in humans have reinforced
metformin's potential role as a therapeutic agent in lung cancer. In the first, Mezzone et
al. showed that diabetic patients with lung cancer previously treated with metformin or
thiazolidinediones had a lower incidence of metastatic disease at the time of diagnosis and a
reduced risk of death compared to those who did not receive the same treatment. More
recently, a retrospective study performed by Tan et al. evaluated the outcomes of three
groups of diabetic patients with NSCLC treated with first line chemotherapy and receiving
various diabetic drugs. In this study, patients treated with chemotherapy with metformin had
superior outcomes compared to those patients treated with chemotherapy with insulin or with
drugs other than metformin (OS, 20 months vs. 13.1 months vs 13.0 months, respectively,
p=0.007). The remarkable activity of this agent in both preclinical and clinical lung cancer
models as well as its low toxicity and tolerability in non diabetic patients warrants further
prospective studies evaluating the therapeutic efficacy with platinum based chemotherapy in
NSCLC.

Metformin's Biological Effects:

A nascent understanding of metformin's biological effect on malignant cells may provide
explanation to the aforementioned preclinical and clinical observations and solidifies the
basis for further clinical exploration. Preclinical models suggest a variety of mechanism
including inhibition of "energy sensing" pathways involved in cellular growth, interference
with the IGF1-insulin axis and blockade of VEGF. Metformin is thought to activate
AMP-activated protein kinase (AMPK), a major sensor of cellular energy levels and a key
enzyme limiting cellular growth during times of cellular stress. Once activated, this enzyme
restricts anabolic processes such as protein, cholesterol and fatty acid synthesis and
inhibits mTOR, a protein kinase responsible for dysregulated growth. MTOR is upregulated in a
variety of tumors, including NSCLC providing rationale to exploit this pathway with
metformin. Secondly, metformin reverses hyperinsulinemia leading to down regulation of
insulin-like growth factors (IGFs). These factors are associated with malignant and
non-malignant tumorogenesis via their aberrant activation of PI3K/Akt pathway, a well known
pathway that contributes to tumorigenesis. Finally, metabolic reprogramming from oxidative
phosphorylation to aerobic glycolysis, known as the Warburg effect, is a hallmark of cancer
cells that constitutes an undisputed advantage in tumor growth. Despite this glycolytic
shift, some malignant cells retain the capacity to continue oxidative phosphorylation for
energy production which may enhance malignant potential. In vivo models indicate that
metformin inhibits mitochondrial complex I thereby suppressing oxidative phosphorylation
which may force cells to engage in survival processes like autophagy leading to eventual cell
death. In summary, metformin's inhibition of tumor cell growth by disrupting both the MTOR
and insulin pathways by AMP kinase activation represents a novel way to treat lung cancer.

Inclusion Criteria:

1. Able to provide written consent and is amenable to compliance with protocol schedules
and testing

2. Patient is > 18 years of age

3. Pathologically proven (either histologic or cytologic) diagnosis of Stage IIIB or IV
non-squamous non-small cell lung cancer

4. No prior, palliative chemotherapy for stage IV lung cancer Patients who have received
adjuvant chemotherapy post surgery for curative intent more than 12 months prior to
development of stage IV disease are allowed.

5. Measurable disease as RECIST criteria 1.1 (Response Evaluation Criteria in Solid
Tumors, Version 1.1)

6. CT Scan of the chest/abdomen/pelvis or PET Scan within 30 days of study entry

7. An MRI of the brain or Head CT Scan with contrast within 30 days of study entry if
clinically indicated

8. ECOG Performance Status 0-2.

9. CBC/differential obtained within 2 weeks prior to registration on study, with adequate
bone marrow function defined as follows:

- Absolute neutrophil count (ANC) >1,500 cells/ul

- Platelets > 100,000 cells/ul

- Hemoglobin > 9.0 g/dl (Note: The use of transfusion or other intervention to
achieve Hgb > g/dl is acceptable.)

10. Serum creatinine < 1.5 x ULN

11. Total bilirubin < 2.0 times the institutional Upper Limit of Normal (ULN)

12. AST and ALT < 3.0 x the ULN

13. Women of childbearing potential must have:

- A negative serum or urine pregnancy test (sensitivity <= 25IU HCG/L) within 14
days prior to the start of study drug administration

- Persons of reproductive potential must agree to use and utilize an adequate
method of contraception throughout treatment and for at least 90 days after study
drug is stopped prior to study enrollment, women of childbearing potential must
be advised of the importance of avoiding pregnancy during trial participation and
the potential risk factors for an unintentional pregnancy.

14. Ability to take oral medication

Exclusion Criteria:

1. The patient has a diagnosis of squamous cell carcinoma. Adenosquamous (mixed)
histologies are allowed

2. The patient has a history of type I or type II diabetes

3. Weight of less than 80% of (IBW) ideal body weight

4. Creatinine clearance less than 45 l/min as calculated by the Cockcroft-Gault equation

5. Known EGFR or ALK mutation in which targeted therapy with erlotinib or crizotinib
would be the standard of care. Those patients whose tissue is not tested or have
insufficient material are eligible

6. The patient is currently taking or has previously taken metformin in the past 6 months

7. The patient has received previous chemotherapy for NSCLC except in instances of
adjuvant therapy post surgical resection more than 12 months prior to enrollment

8. The patient has undergone major surgery within four weeks prior to randomization.

9. The patient has undergone palliative radiation (chest, brain) to tumor sites within
two weeks of randomization (except palliative radiation to the bone which can be
within one week

10. Uncontrolled (untreated) brain metastasis.

11. Patient who has NCI-CTCAE Version 4 Grade >= 2 diarrhea

12. That patient has clinically relevant CAD or uncontrolled CHF

13. The patient has ongoing or active infection (requiring antibiotics) that would limit
the administration of chemotherapy including active TB. HIV is allowed in this study

14. The patient has a history of neurological or psychological disorder that may interfere
with the compliance of the protocol

15. Women who are unwilling or unable to use an acceptable method to avoid pregnancy for
the entire study period and for at least 4 weeks after cessation of study drug, or
have a positive pregnancy test at baseline, or are pregnant or breastfeeding