Pharma Strategy Blog

Commentary on Pharma & Biotech Oncology / Hematology New Product Development

Posts tagged ‘AKT’

Update on RAF resistance in metastatic melanoma

“RAF inhibitors (vemurafenib and dabrafenib) have profound clinical activity in patients with BRAF-mutant melanoma, but their therapeutic effects are limited by the emergence of drug resistance.”

Solit and Rosen (2014)

For today’s post on Science Fridays, I wanted to take a look at an overview paper, published in Cancer Discovery, from two researchers in the metastatic melanoma field who have been looking at multiple mechanisms of resistance.  It’s an important topic because while we have seen incremental improvements in outcomes for this disease, the 5-year survival rate is still rather poor with only 10–20% of metastatic patients still alive by then.  This is not to disparage the efforts of scientists, clinicians or companies working in this space, far from it, but there is is clearly a need for new therapies, strategies and combinations, given the high unmet medical need that exists.

We still have a long way to go in moving the survival needle dramatically.

It wasn’t until I searched for related blog posts to link to this one that I realised how much we have already covered on this topic! Regular readers will recall discussions here on PSB on various combinations such as:

  1. RAF + MEK inhibitors (downstream resistance)
  2. RAF + PI3K-AKT-mTOR inhibitors (cross resistance)
  3. RAF + CTLA–4 checkpoint inhibitors (anti-tumour immunity)

to name a few examples.

We have seen that adding a MEK inhibitor to dabrafenib e.g. trametinib can overcome resistance temporarily and add a few extra months before the resistance sets in again. Similarly for PI3K inhibitors tested to date. Adding ipilimumab, an anti-CTLA–4 checkpoint inhibitor held much promise, but the combination was abandoned with the emergence of unexpected liver toxicity.

Results thus far suggest that something else is acting as an escape route, thereby enabling the tumour to continue driving oncogenic addiction to BRAF.

The $64K questions are what is happening and what can we do about it?

We also need to remember that clinical research advances piecemeal based on evidence from preclinical reseach, so we see the logical evolution of BRAF monotherapy -> combos with downstream (MEK) or upstream (NRAS) targets in same pathway -> combos with diagonal (PI3K) pathways etc.

What Solit and Rosen have done is put a nice summary together of the state of play in this disease and the paper (see References below) is well worth reading.

Their main assertion is interesting, namely:

“The common feature of each of these mechanisms of resistance is that they result in activation of ERK signaling that is insensitive to the RAF inhibitor. Thus, RAF inhibitor resistance is often associated with maintenance of activation of the oncogene-driven pathway.”

Two recent papers are cited in support of this theory from Shi et al., (2014) and Allen et al., (2014) – see References below for additional background reading. Both studies used patient samples to look at clonal evolution and the genetic landscape in advanced melanoma. It’s actually quite amazing what unbiased exome sequencing can uncover at the molecular level, not least are the development of new mutations and other functional alterations.

The Shi et al., (2014) study was briefly summarised by Solit and Rosen:

“Multiple biopsies were obtained at different times or from disparate locations from several patients, and more than a single lesion in the ERK pathway was identifi ed in multiple patients typically within
different tumor biopsies.”

They went to note:

“A detailed phylogenetic analysis of multiple progressive lesions from a subset of these patients suggested branching evolution of tumors in which the development of genetic diversity was not linearly associated with time.”

Previously, a case report found distinct mechanisms of BRAF inhibitor resistance were present in two different progressing lesions from a single patient, so the work of Shi et al., (2014) is consistent with this finding. It blows my mind that different lesions in the same patient might behave completely differently though – imagine trying to devise an appropriate and effective clinical strategy in these cases?!

Allen et al’s (2014) work also involved whole exome sequencing (WES) from patient samples:

“WES was performed on paired pretreatment and progression samples collected from 45 patients, of whom 14 developed resistance soon after initiation of therapy (within 12 weeks). They also detected several resistance mechanisms that had been previously identified to confer RAF inhibitor resistance, including mutations in NRAS , MAP2K1, and NF1 and BRAF amplification.”

A third important study in this area from Wagle et al., (2014) adds to the weight of evidence with new mutations developing. Solit and Rosen continued the story:

“Consistent with the preclinical studies highlighted above demonstrating that MEK1 and MEK2
mutations can confer RAF and MEK inhibitor resistance, a MEK2 Q60P mutation was identifi ed in 1 of 5 patients studied. Of greater surprise to the investigators, one patient had a BRAF splice variant lacking exons 2–10 and a second patient had BRAF amplification.”

By now, you can see the sheer variety of changes and adaptations taking place in different studies around the world in some of the top melanoma labs. What do they have in common though?

“One hypothesis to explain this result is that increased abundance of the oncogenic driver (in this case BRAF) in response to prolonged drug treatment results in increased flux through the ERK pathway and restoration of ERK activity above the threshold required for inhibition of cell proliferation.”

The next challenge is to figure how we can approach better therapeutic index and shutting down of the pathways?

“The results suggest that the early adaptive response of BRAF -mutant cells to ERK pathway inhibition may promote the selection of resistant clones that harbor additional genomic events that
confer higher levels of RAF inhibitor resistance. The data also support combinatorial approaches that attenuate the adaptive response, including the addition of a PI3K or AKT inhibitor to the RAF and MEK (or ERK) inhibitor combination.”

The problem with this approach though, is that the neither the BRAF nor PI3K inhibitors have been able to reach or go beyond the single agent dosing schedules:

“As previous attempts to combine MAPK and PI3K pathway inhibitors have been limited by overlapping toxicities, upfront testing of intermittent treatment schedules should be considered.”

This is the also approach that Das Thakur suggested in her work presented at AACR last year, and subsequently published in Nature, to delay the development of resistance to vemurafenib.

I do think this one area where we may well see new trials evolve in advanced melanoma, so we will have to wait for new data before we can see if the strategy is successful at delaying the emergence of resistant clones. It is good to see the evolution of solid preclinical and translational evidence from patient biopsies helping to inform future clinical trial strategies.

In the meantime, the next major milestone I’m waiting for is on Roche/Genentech’s MEK inhibitor, cobimetinib (GDC–0973), which is due to report combination data with vemurafenib (continuous dosing) later this year. It will be interesting to see if this inhibits MEK more completely than trametinib and whether the combination has a better initial outcome than dabrafenib plus trametinib, which added about two to three months of extra survival over dabrafenib alone.

References:

ResearchBlogging.orgSolit DB, & Rosen N (2014). Towards a Unified Model of RAF Inhibitor Resistance. Cancer discovery, 4 (1), 27–30 PMID: 24402945

Shi H, Hugo W, Kong X, Hong A, Koya RC, Moriceau G, Chodon T, Guo R, Johnson DB, Dahlman KB, Kelley MC, Kefford RF, Chmielowski B, Glaspy JA, Sosman JA, van Baren N, Long GV, Ribas A, & Lo RS (2014). Acquired Resistance and Clonal Evolution in Melanoma during BRAF Inhibitor Therapy. Cancer discovery, 4 (1), 80–93 PMID: 24265155

Van Allen EM, Wagle N, Sucker A, Treacy DJ, Johannessen CM, Goetz EM, Place CS, Taylor-Weiner A, Whittaker S, Kryukov GV, Hodis E, Rosenberg M, McKenna A, Cibulskis K, Farlow D, Zimmer L, Hillen U, Gutzmer R, Goldinger SM, Ugurel S, Gogas HJ, Egberts F, Berking C, Trefzer U, Loquai C, Weide B, Hassel JC, Gabriel SB, Carter SL, Getz G, Garraway LA, Schadendorf D, & Dermatologic Cooperative Oncology Group of Germany (DeCOG) (2014). The Genetic Landscape of Clinical Resistance to RAF Inhibition in Metastatic Melanoma. Cancer discovery, 4 (1), 94–109 PMID: 24265153

Wagle N, Van Allen EM, Treacy DJ, Frederick DT, Cooper ZA, Taylor-Weiner A, Rosenberg M, Goetz EM, Sullivan RJ, Farlow DN, Friedrich DC, Anderka K, Perrin D, Johannessen CM, McKenna A, Cibulskis K, Kryukov G, Hodis E, Lawrence DP, Fisher S, Getz G, Gabriel SB, Carter SL, Flaherty KT, Wargo JA, & Garraway LA (2014). MAP Kinase Pathway Alterations in BRAF-Mutant Melanoma Patients with Acquired Resistance to Combined RAF/MEK Inhibition. Cancer discovery, 4 (1), 61–8 PMID: 24265154

Das Thakur M, Salangsang F, Landman AS, Sellers WR, Pryer NK, Levesque MP, Dummer R, McMahon M, & Stuart DD (2013). Modelling vemurafenib resistance in melanoma reveals a strategy to forestall drug resistance. Nature, 494 (7436), 251–5 PMID: 23302800

Can Merck make an impact in pancreatic cancer?

There were a number of interesting posters at the AACR-NCI-EORTC Molecular Targets meeting today. Specifically, two on metastatic pancreatic cancer caught my eye. You can read about the other one on Millennium/Takeda’s ADC MLN0264 here.

This is an area of high unmet medical need with the fourth highest number of cancer deaths in the US and a median survival of 10 months or less. Even with improvements in the standard of care, it still remains a miserable cancer to get.

Many of you will be aware that KRAS is mutated in 90% of pancreatic cancer cases. As Dr Barry Nelkin (Johns Hopkins) noted today,

“The good news is that we know the target, the bad news is that we haven’t been able to hit it… Yet.”

We do know that KRAS activates three major signaling pathways, namely:
1. RAF/MEK/ERK
2. PI3K/AKT/mTOR
3. RAL.

Oddly, we have a plethora of inhibitors for the first two, but not for RAL. Unfortunately, RAL signaling is critically important in pancreatic cancer.

Dr Nelkin observed that it has been shown that inhibition with CDK5 resulted in the loss of RAL activity in pancreatic cancer cells, as well as reduction in their ability to form tumours and metastasize in vivo. Interestingly, addition of PI3K or MEK inhibitors further decreased the cells transformation.

Dinaciclib (Merck), an inhibitor of CDK1,2,5 and 9 was shown to block RAL activity in pancreatic cancer cells as well as inhibiting xenograft growth and metastasis. In other words, RAL is now druggable!

The researchers at Johns Hopkins showed some nice preclinical evidence that tumour size was reduced in pancreatic cell lines when dinaciclib was combined with either an Akt inhibitor (MK-2206) or an ERK inhibitor (SCH772984).

These results therefore provided a solid rationale for combining a CDK5 inhibitor with one from the PI3K/Akt/mTOR or RAF/MEK/ERK pathways in the clinic. A phase I combination trial of dinaciclib plus MK-2206 is currently enrolling at four centres, including three in the US.

Overall:

I think this is a most interesting trial with a solid rationale that is well worth evaluating in advanced pancreatic cancer. The story though, gets a little interesting. I tweeted to Dr Nelkin’s colleague, Dr Anirban Mitra, who is now Professor of pancreatic cancer research at MD Anderson and learned something rather surprising:

Cytocidal for the uninitiated means they are seeing pancreatic cancer cells being killed.  It’s a short word but it means a lot.

At the poster today, Dr Nelkin confirmed that he had heard rumours to this effect, but stated Merck had at least committed to finishing the phase I trial. It’s not immediately obvious why MK-2206 might be discontinued or on the chopping block, but who knows what will happen if the phase I data turns out to be stunning.  I for one, sincerely hope that they are.

Other companies with a CDK5 and a PI3K/AKT/mTOR inhibitor in their pipeline would do well to watch out for the readout of these results – they could be very interesting indeed.

What’s hot at ECCO 2013 – Preview of late breakers and PI3K abstracts

By on September 17, 2013

Next I’ll be off to the European Cancer Congress (ECC) in Amsterdam. This meeting alternates each year between ECCO and ESMO hosting the event at a different European city.

The last couple of years have seen some nice data that missed the ASCO deadline, other years can bring an update of the already familiar ASCO data. I suspect that this year will be one of those events, with updated PD-1 and PD-L1 data.

If you missed my colleague Pieter Droppert’s ECCO highlights yesterday, you can catch them here, including details of the iPad app and abstracts.

In addition, there were other abstracts of interest that caught my eye, including some solid late breakers:

1. T-DM1 for HER2-positive metastatic breast cancer (MBC): Primary results from TH3RESA, a phase 3 study of T-DM1 vs treatment of physician’s choice. H. Wildiers (Belgium) et al.

The study looks at advanced disease in patients who had received at least two prior regimens. This analysis looks like an interim one, given the full study timeframe is over 3.5 years. I’m particularly curious what the physician choices were to compete with Kadcyla and what the 1 year survival curves look like. It’s a wee bit early to hope that they might separate already, certainly I hope they do!

2. Evaluation of everolimus (EVE) in HER2+ advanced breast cancer (BC) with activated PI3K/mTOR pathway: Exploratory biomarker observations from the BOLERO-3 trial. G. Jerusalem (Belgium) et al.

Originally, I thought this had been presented at ASCO, but the biomarker abstract I found actually referred to BOLERO-2, where they noted that “efficacy was greater in patients with low PI3K expression”, which is an odd finding. The BOLERO-3 data from ASCO presented the initial phase III data for the combination of trastuzumab, vinorelbine and everolimus vs trastuzumab and vinorelbine alone in trastuzumab resistant HER2+ advanced breast cancer. This should be an interesting presentation worth attending.

3. FLT1 gene variation as a major determinant of recurrence in stage I-III non-small cell lung cancer. F. Innocenti (USA) et al.

Many of us familiar with FLT3 in leukemia, but FLT1 is an interesting concept with very little data (or drugs) out there. I will be curious to see if this is a druggable target and where this approach might lead.

4. Pooled analysis of long-term survival data from phase II and phase III trials of ipilimumab in metastatic or locally advanced, unresectable melanoma. F. Hodi (USA) et al.

If you look at five year survival curves for advanced melanoma in the literature, it’s historically around 20% or so when patients have received IL-2, which is where I would expect ipilimumab to be. Trials with DTIC have shown a much lower rate, at around 8-9%. Not every patient is suitable for IL-2 though, so we may be seeing similar survival rates irrespective of the immunotherapy given, but with very different safety profiles.

One of my favourite cancer pathways is PI3K-AKT-mTOR. It’s dysregulated in some 80% of cancers yet we haven’t really seen a major breakthrough with these agents in solid tumours outside of the stunning Afinitor data in relapsed ER+ metastatic breast cancer from the BOLERO-2 study presented at ECCO in Stockholm two years ago.

There are many different permutations out there from single to dual inhibitors and also specific isoforms of alpha, beta, delta and gamma.

One of the challenges with targeting PI3K is that it activates feedback loops. Thus inhibiting PI3K in advanced prostate cancer activates the androgen receptor, while single agent use in advanced breast cancer can lead to activation of HER3. In addition, there have been mixed results with biomarkers and specific mutations/tumour suppressors to date such as PIK3CA and PTEN. This increases the complexity tremendously and therefore speaks to more careful trial selection based on inclusion criteria and also logical combinations to try and shut down the compensatory pathway.

I was therefore pleased to see a few trials reporting early phase I/II data in this vein:

P017: Evaluation of tolerability and anti-tumor activity of GDC-0032, a PI3K inhibitor with enhanced activity against PIK3CA mutant tumors, administered to patients with advanced solid tumors.

D. Juric, J.R. Infante, I.E. Krop, C. Kurkjian, M.R. Patel, R.A. Graham, T.R. Wilson, J.Y. Hsu, J. Baselga, D.D. Von Hoff

According to the abstract:

“GDC-0032 is an orally bioavailable, potent, and selective inhibitor of Class I PI3K alpha, delta, and gamma isoforms, with 30-fold less inhibition of the PI3K beta isoform relative to the PI3K alpha isoform.”

Several confirmed partial responses have been reported and further trials will continue:

“GDC-0032 is a next-generation PI3K inhibitor with promising anti-tumor activity observed in patients with PIK3CA mutant tumors. GDC-0032 is being investigated in combination with endocrine therapies such as letrozole and fulvestrant for patients with hormone receptor-positive breast cancer.”

P079: Hyperglycemia in patients treated with the pan-PI3K inhibitor buparlisib (BKM120): characterization, management, and assessment for pharmacodynamics

A. Azaro, J. Rodon, J.F. Vansteenkiste, Y. Ando, T. Doi, D. Mills, C. Sarr, E. Di Tomaso, C. Massacesi, R.W. Naumann

Source: Novartis

Source: Novartis

BKM120 is also an oral pan PI3K inhibitor that does not target mTOR. Aside from the activation feedback loop effects mentioned earlier, PI3K plays a key role in glucose homeostasis. A number of earlier trials with different PI3K and mTOR inhibitors have reported hyperglycemia as a class effect although they have varied in the degree to which the event occurred.

This study highlights the importance of a potential pharmacodynamic marker (C-peptide) in assessing the insulin response and I’m looking forward to seeing more detail in the poster.

P061: Factors predisposing to development of hyperglycaemia in phase 1 studies involving PI3K, mTOR, AKT and mTORC1 and mTORC2 inhibitors

M. Wong, K.H. Khan, K. Rihawi, S. Bodla, B. Amin, K. Shah, D. Morganstein, S.B. Kaye, U. Banerji, L.R. Molife

Related to the topic of hyperglycemia, the Royal Marsden mined their database for PI3K-mTOR trials and looked at factors that might influence the presence of the glucose spike in order to essentially try and predict which patients were more at risk and improve management. While most patients did not require intervention, but in those that did, metformin and insulin were usually preferred. Interestingly, the main factor emerging in this retrospective study was a prior history of diabetes, which is not totally unsurprising. It will be useful to see if these results can be validated in prospective future trials.

P227: Anti-tumour efficacy of the PI3K inhibitor GDC0941, the dual PI3K/mTOR inhibitor GDC0980 and the MEK inhibitor GDC0973 as single agents and in combination in endometrial carcinomas

O. Aslan, A.M. Farrelly, B. Stordal, B.T. Hennessy

Much has be written about the potential for a PI3K and MEK combination in different tumour types, but so far they haven’t proven to be the home run many of us hoped for.

This preclinical paper looks at cell lines to explore potential targets and synergies in endometrial cancer (EC). They concluded,

“Our data suggest that the mutational status of PIK3CA, PTEN and KRAS can be used as biomarkers to select patients for PI3K and RAS/RAF-targeted therapies. Further, the combinations of the PI3K inhibitors GDC0941 and GDC0980 with the MEK inhibitor GDC0973 are promising approaches for the treatment of patients with PIK3CA, PTEN and KRAS-mutated EC.”

Translating data from simple cell lines to complex human bodies does not always predict response given the variable responses seen from patients with mutations and tumour suppressors in clinical trials. I think it will take a while to tease out what defines and predicts a response in each tumour type much in the same way we saw different effects in advanced melanoma when targeting BRAF with sorafenib versus BRAF V600E with vemurafenib or dabrafenib. The devil is in the details.

And finally, an oral presentation with a very different focus in the PI3K related field that I’m really looking forward to hearing:

#1859 PI3KCA mutations and correlation with pCR in the NeoALTTO trial (BIG 01-06)

J. Baselga, I. Majewski, P.G. Nuciforo, H. Eidtmann, E. Holmes, C. Sotiriou, D. Fumagalli, M.C. Diaz Delgado, M. Piccart-Gebhart, R. Bernards

The authors evaluated:

“The influence of PI3K pathway mutations (PIK3CA, KRAS, BRAF, AKT1) on sensitivity to trastuzumab (T), lapatinib (L), or both agents (L+T) in combination in early-stage HER2-positive breast cancer patients enrolled the neoALTTO trial.”

The goal here is to see the presence of any of the mutations were more likely to lead to resistance and enable better selection of therapy for patients. I will update on this study after the presentation.

More detailed posts and synopses will continue from the meeting itself on Biotech Strategy Blog, where we’ll be sharing our insights and analysis daily.

ASCO 2012 Preview – Highlights of key data emerging from Chicago

By on May 15, 2012

It’s that time of the year again where we cogitate and contemplate on what might be hot at the annual meeting of the American Society of Clinical Oncology (ASCO) before the abstracts are available (they’re released online tomorrow at 6pm ET).

This year, while interesting early data from up and coming small biotechs is likely to be eagerly presented in poster sessions, the focus is more likely going to be on big Pharma with various phase III and also late phase II trials that are due to report data.  Unfortunately, not all of these will produce overwhelmingly positive results though!

What I’m most interested is things that shift the needle meaningfully  in terms of survival by six months or more, as we saw from the recent BOLERO2 and CLEOPATRA trials in ER+ and HER2+ breast cancer.  There are plenty of agents that offer minor or incremental improvements (colon cancer has long suffered from that syndrome, sadly), but let’s be honest – most of us get excited by the possibility of major shifts in survival.

Please note that I’ve mostly selected some promising agents in development that might achieve that effect, explained why they are different and focused on new data/drugs rather than rehash what I call the ‘middlings’ i.e. minor upgrades to the standard of care.

Without much further ado, here are my ASCO preview highlights for 2012:

Please do check back during the convention both here on PSB, and also on Biotech Strategy, for reports and analysis as the interesting data emerges at ASCO.

If you have any comments or thoughts, please do share them below…

4 Comments

Will combining custirsen and MDV3100 reduce resistance in advanced prostate cancer?

By on February 26, 2012

EAU-2012-Congress-Paris-View-of-Eiffel-Tower-By-NIght

Sights of 2012 EAU Congress

Greetings from the European Association of Urology (EAU) congress in Paris. Despite the grey drizzle typical of Europe in winter, this is actually quite an interesting meeting with lots of poster presentations.

One poster that caught my eye yesterday was from Martin Gleave’s group on clusterin knockdown synergising MDV3100 activity. Previously, we discussed on this blog how inhibiting clusterin with custirsen (OGX-011) potentiated docetaxel. At the AUA meeting last year, the issue of whether the same would happen with MDV3100 was suggested, as you can see in the short video blog.

Clusterin is also known as testosterone-repressed prostate message-2 or TRMP-2, and has been shown by Miyake et al., (2000) to be important in advanced prostate cancer. This is because the treatment of choice in hormone-sensitive disease, androgen deprivation therapy (ADT), can lead to clusterin upregulation, thereby causing castrate resistance.

The group’s latest study at EAU looked at whether clusterin knockdown sensitised MDV3100 activity and evaluated potential mechanisms for how this might work.

The results showed that:

  1. Neither MDV3100 or custirsen alone affected AR levels, but in combination, the AR protein levels were reduced.
  2. The combination synergistically suppressed LNCaP (human prostate cancer cell lines) in vitro and in vivo compared to monotherapy with either alone.
  3. Inhibition of the AR has been shown to activate the PI3K-Akt pathway, but the combination prevented this from occurring.
  4. Dual treatment also increased AR instability via decreased levels of the AR chaperone, FKBP52.
  5. AR degradation occurred with combination therapy via the proteasome, leading to synergistic repression of AR transcription.

While these data offer a very nice and logical preclinical rationale for considering a combination of MDV3100 and custirsen to overcome castrate resistance in advanced disease, we also need to see clinical evidence in advanced prostate cancer before getting too excited. I like the idea scientifically but Oncogenex, the manufacturers of custirsen, have not exactly been swift at moving their previous trials along, as Luke Timmerman noted his Xconomy article last year.

Ultimately, the proof is always in the (clinical) pudding.

References:

ResearchBlogging.orgMiyake H, Nelson C, Rennie PS, & Gleave ME (2000). Testosterone-repressed prostate message-2 is an antiapoptotic gene involved in progression to androgen independence in prostate cancer. Cancer research, 60 (1), 170-6 PMID: 10646870

2 Comments

EGFR activates mTORC2 in glioblastoma – a potential role for new therapies

By on December 1, 2011

One of the great things about following the American Association for Cancer Research (AACR) on Twitter, is that they regularly share technical open access articles from their journals for scientists to read.  Of course, many will have access through their institution subscription, but there are also probably quite a few interested community oncologists and scientists like me that don’t. The idea of sharing some of their really important scientific research with the broader public is a great one – a little bit of goodwill goes a long way and furthers their cause too.

Yesterday, AACR kindly tweeted and shared a fascinating paper (see references below for open access to all the articles) on how EGFR signaling in glioblastoma (an aggressive form of brain cancer) activates the mTOR pathway, specifically mTORC2, and is partially suppressed by PTEN:

EGFRmTOR
Source: Tanaka et al., (2011)

We know that mTOR and it’s upstream relative, PI3K, are frequently dysregulated in cancer and may also lead to resistance to treatment with some therapies, such as aromatase inhibitors in breast and other cancers. This is also true in glioblastoma, where chemotherapies such as temozolamide are often used, as the authors noted:

“mTORC2 signaling promotes GBM growth and survival and activates NF-κB. Importantly, this mTORC2–NF-κB pathway renders GBM cells and tumors resistant to chemotherapy in a manner independent of Akt.”

One of the challenges though, is elucidating the mechanism behind mTOR activation:

“The mechanisms of mTORC2 activation are not well understood. Growth factor signaling through PI3K, potentially through enhanced association with ribosomes, and up-regulation of mTORC2 regulatory subunits have been proposed as mechanisms of mTORC2 activation.”

Recently, Clohessy et al., (2008) observed that mTORC1 inhibition was not sufficient to block GBM growth, so this new research took a different approach and focused on asking the question of whether oncogenic EGFR affects mTORC2. To test this hypothesis, they used GBM derived cell lines that represent the most common genetic events driving GBM i.e. PTEN loss with EGFR overexpression or activating mutation (EGFRvIII) present or absent. It should be noted that a good marker of mTORC2 activity is the phosphorylation of AKT S473, although SGK1 is also turning out to be a good biomarker of response.

What did they find?

The paper (open access) is well worth reading, but to summarise, here are some of the key findings from this well thought out research:

  • mTORC2 signaling promotes GBM growth and survival
  • EGFRvIII activates NF-kB through mTORC2
  • mTORC1 inhibition alone could not suppress NF-κB activation in GBM cells
  • mTORC2 mediates EGFRviii-dependent cisplatin resistance through NF-kB, independently of Akt
  • mTORC2 inhibition reverses cisplatin resistance in xenograft tumours
  • mTORC2 signaling is hyperactivated and associated with NF-kB and phospho-EGFR in the majority of clinical GBM samples

What stood out for me in their series of experiments and comprehensive analysis was that:

“Elevated phosphorylation of EGFR (Y1068) and Akt (S473) was detected in 44% and 77% of GBMs, respectively. These numbers are consistent with the independent findings of EGFR mutation and/or amplification in 45% and PI3K pathway–activating mutations in 87% of GBMs, reported in the Cancer Genome Atlas studies.”

What do these results all mean?

Looking at question regarding the mechanism underlying mTORC2 activation and its relationship with EGFR was poorly understood, this paper clearly showed that mTORC2 activation is a common event in GBM, including tumors harbouring EGFR-activating lesions. But what was particularly interesting was the finding that EGFRvIII was significantly more potent than wild-type EGFR in promoting mTORC2 activity. This is consistent with previous work from Huang et al., (2007), who found that:

“EGFRvIII preferentially activates PI3K signaling despite lower levels of receptor phosphorylation, leading to differential activation of downstream effectors.”

One outstanding question that has puzzled many researchers is what is the mechanism of rapamycin (mTOR) resistance? There are some clues in this research:

“Here we demonstrated that rapamycin (or genetic mTORC1 inhibition by raptor knockdown) promoted Akt S473 and NDRG1 T346 phosphorylation; this feedback activation could be suppressed by mTORC2 inhibition.”

They also looked at a patient sample to determine if there were any hints for further translational research:

“In a clinical sample from a GBM patient analyzed before and 10 days after treatment with rapamycin, mTORC2 signaling was elevated concomitant with significant mTORC1 inhibition, as measured by decreased S6 phosphorylation.”

This is important because to date, based on much of the data that has emerged from mTOR and PI3K inhibitors we have seen that single agent therapy often leads to either stable disease or low response rates, so the question is how can we improve this by understanding the mechanisms of resistance better in order to direct future combination approaches (as opposed to single agent studies) logically:

“These data suggest the possibility that failure to suppress mTORC2 signaling, including NF-κB signaling, may underlie resistance to rapamycin and the poor clinical outcome associated with it in some patients with GBM.”

This is a crucial finding because some early mTOR inhibitors such as rapamycin target mTORC1 effectively, but are weak inhibitors of mTORC2. The new generation of inhibitors may address this issue better and shut down the mTOR pathway more effectively, although that may not be enough on it own.

Clearly, future research studies will be needed to better understand the potential role of mTORC2/NF-κB signaling in mediating resistance to treatment in GBM:

“The results reported here provide a potential mechanism for mutant EGFR-mediated NF-kB activation in GBM and other types of cancer. The results also suggest that EGFR tyrosine kinase inhibitor resistance could also potentially be abrogated by targeting mTORC2-mediated NF-kB activation.”

So far this is a good start, but we still have a long way to go. There are a number of mTOR and PI3K inhibitors in development for the treatment of GBM – I’m looking forward to seeing the results of those trials and learning which combinations and lines of therapy might see the best results with mTOR inhibitors. Hopefully, there might be some early readouts at ASCO next June.

References:

ResearchBlogging.orgTanaka, K., Babic, I., Nathanson, D., Akhavan, D., Guo, D., Gini, B., Dang, J., Zhu, S., Yang, H., De Jesus, J., Amzajerdi, A., Zhang, Y., Dibble, C., Dan, H., Rinkenbaugh, A., Yong, W., Vinters, H., Gera, J., Cavenee, W., Cloughesy, T., Manning, B., Baldwin, A., & Mischel, P. (2011). Oncogenic EGFR Signaling Activates an mTORC2-NF- B Pathway That Promotes Chemotherapy Resistance Cancer Discovery, 1 (6), 524-538 DOI: 10.1158/2159-8290.CD-11-0124

Cloughesy TF, Yoshimoto K, Nghiemphu P, Brown K, Dang J, Zhu S, Hsueh T, Chen Y, Wang W, Youngkin D, Liau L, Martin N, Becker D, Bergsneider M, Lai A, Green R, Oglesby T, Koleto M, Trent J, Horvath S, Mischel PS, Mellinghoff IK, & Sawyers CL (2008). Antitumor activity of rapamycin in a Phase I trial for patients with recurrent PTEN-deficient glioblastoma. PLoS medicine, 5 (1) PMID: 18215105

Huang, P., Mukasa, A., Bonavia, R., Flynn, R., Brewer, Z., Cavenee, W., Furnari, F., & White, F. (2007). Quantitative analysis of EGFRvIII cellular signaling networks reveals a combinatorial therapeutic strategy for glioblastoma Proceedings of the National Academy of Sciences, 104 (31), 12867-12872 DOI: 10.1073/pnas.0705158104

Update on EGFR resistance in lung cancer: new directions

By on October 25, 2011

This morning I was reading a fascinating paper on lung cancer and one of my favourite proteins, CRKL, from the group of prolific lung researchers at Mass General, Dana Farber, MIT and the Broad Institute in Boston:

“Over-expression of CRKL in immortalized human airway epithelial cells promoted anchorage-independent growth and tumorigenicity. Oncogenic CRKL activates the SOS1-RAS-RAF-ERK and SRC-C3G-RAP1 pathways. Suppression of CRKL in NSCLC cells that harbor CRKL amplifications induced cell death.”

Cheung et al., (2011)

We also know that one of the mechanisms of resistance to gefitinib is over-expression of CRKL in EGFR-mutant cells by activating ERK and AKT signaling.

What was interesting about this research was the observation:

“We identified CRKL amplification in an EGFR inhibitor-treated lung adenocarcinoma that was not present prior to treatment.”

Emphasis mine.

We do know that:

  • Adaptive resistance to treatment is a common problem with kinase inhibitors
  • Some lung cancer tumours acquire the T790M mutation, which is known to confer resistance to EGFR therapies
  • Several groups have also reported other known resistance mechanisms may also occur with the EGFR T790M mutation, including MET amplification and CTNNB1 (β-catenin) mutations.

Cheung et al., (2011) tested to see if the PI3K-AKT pathway was specifically involved with CRKL resistance:

“We examined whether treatment with the PI3K inhibitor GDC-0941 suppressed growth of CRKL–over-expressing HCC827 cells in response to gefitinib. Cells were exposed to GDC-0941 alone or in combination with gefitinib. Combined treatment with GDC-0941 and gefitinib resulted in a substantial decrease in the relative proliferation of CRKL–over-expressing HCC827 cells compared to gefitinib treatment alone.”

The answer was yes, activation of PI3K-AKT signalling contributes to CRKL-induced EGFR inhibitor resistance.

It would therefore be very interesting to see what happens in the clinic to a subset of lung cancer patients with CRKL amplification who are treated with an EGFR and PI3K inhibitor to see if this reduces resistance to treatment and improves outcomes. Trials with the combination are indeed ongoing, although I think they are in a more general population of patients with EGFR driven lung cancer. Based on these findings, a subset analysis might prove to be rather instructive here.

What do these results mean?

This study strongly suggests that CRKL may well be a valid therapeutic target:

“These observations show that CRKL over-expression induces cell transformation, credential CRKL as a therapeutic target for a subset of NSCLC that harbor CRKL amplifications, and implicate CRKL as an additional mechanism of resistance to EGFR-directed therapy.”

“Although CRKL amplifications occur in a relatively small fraction of NSCLC, the finding that a similar fraction of NSCLC with translocations involving ALK respond to treatment with crizotinib indicates that targeting genetic alterations present even in a subset of NSCLC may have clinical importance.”

The general idea that CRKL could act as an oncogene in other cancers with CRKL amplifications is also an intriguing idea that needs be explored further.

The paper is very well written and worth checking out for those interested in EGFR mutations, resistance to therapy and development of new therapies.

References:

ResearchBlogging.orgCheung, H., Du, J., Boehm, J., He, F., Weir, B., Wang, X., Butaney, M., Sequist, L., Luo, B., Engelman, J., Root, D., Meyerson, M., Golub, T., Janne, P., & Hahn, W. (2011). Amplification of CRKL induces transformation and EGFR inhibitor resistance in human non small cell lung cancers Cancer Discovery DOI: 10.1158/2159-8290.CD-11-0046

Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, Park JO, Lindeman N, Gale CM, Zhao X, Christensen J, Kosaka T, Holmes AJ, Rogers AM, Cappuzzo F, Mok T, Lee C, Johnson BE, Cantley LC, & Jänne PA (2007). MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science (New York, N.Y.), 316 (5827), 1039-43 PMID: 17463250

Sequist, L., Waltman, B., Dias-Santagata, D., Digumarthy, S., Turke, A., Fidias, P., Bergethon, K., Shaw, A., Gettinger, S., Cosper, A., Akhavanfard, S., Heist, R., Temel, J., Christensen, J., Wain, J., Lynch, T., Vernovsky, K., Mark, E., Lanuti, M., Iafrate, A., Mino-Kenudson, M., & Engelman, J. (2011). Genotypic and Histological Evolution of Lung Cancers Acquiring Resistance to EGFR Inhibitors Science Translational Medicine, 3 (75), 75-75 DOI: 10.1126/scitranslmed.3002003

3 Comments

Update on MEK and AKT inhibition in RAS driven cancers

By on October 3, 2011

A couple of interesting developments have emerged over the last week with AKT and MEK inhibitors, specifically Merck’s MK-2206 and AstraZeneca/Array’s AZD6244, that are well worth discussing.

  1. At the ECCO/EMCC meeting in Stockholm last Tuesday, Johann De Bono discussed the combination data for MK-2206 and AZD6244 in KRAS driven colorectal cancer.
  2. Later the same week, Array Biopharma announced the initial results from a randomized phase II placebo-controlled study that compared the efficacy of selumetinib (AZD6244/ARRY-886) in combination with docetaxel compared to docetaxel alone in the second-line treatment of patients (n=87) with KRAS-mutant, locally advanced or metastatic non-small cell lung cancer (NSCLC).

Now, to be clear, I like the concept of AKT and MEK inhibitors, especially in select combinations, but the key thing here is the right combinations in the right context.

Let’s take a look at the lung cancer KRAS data first. One of the challenges I have with this approach, is that we’ve know for a while that BRAF and KRAS driven cancers behave rather differently according to Wee et al., (2009):

“Previous studies have found that whereas BRAF mutant cancers are highly sensitive to MEK inhibition, RAS mutant cancers exhibit a more variable response.”

Variable response is not an encouraging phrase when planning clinical trials!

Let’s take a look at the pathway itself:

We can immediately see that MEK is downstream of RAS, meaning that even if we target MEK, unfortunately RAS and KRAS is still largely untouched upstream. This is important to remember when considering the actual results later.

The other key factor to consider is what are the adaptive resistance pathways that might evolve as a result of treatment with a MEK inhibitor? In an ideal world, logical combinations would be tested that target both the primary driving mutation or aberration, as well as the adaptive resistance, to try and shut down the pathway more completely than targeting either alone. Another key question that needs to be addressed is what is driving the KRAS aberrant activity in the first place?

We’ve discussed MEK numerous times here on PSB, but the Wee et al., (2009) MEK paper stands out in particular. They identified a critical resistance pathway to MEK inhibition, namely PI3K. Although we discussed this originally in the context of BRAF driven tumours such as melanoma, it is well worth discussing again here in regards to KRAS driven tumours given a MEK inhibitor is being tested.

They observed that:

“Activating mutations in PIK3CA reduce the sensitivity to MEK inhibition, whereas PTEN mutations seem to cause complete resistance.”

It isn’t clear from the Array press release whether any of the patients with NSCLC exhibited PIK3CA mutations or loss of PTEN, but they definiely do occur in this disease. It will be interesting to see of more meta data is available at the forthcoming AACR Molecular Targets meeting next month.

I’m not a big fan of chemotherapy plus a single targeted agent, because as you can see from the evidence above, the pathway is not being shut down by one targeted agent and resistance is not being addressed at all. The chances of such a combination working (by that I mean increasing overall survival), I think would be fairly low.

According to the press release, the study did not see a significant improvement in overall survival (OS) but did show an encouraging response in the form of progression free survival (PFS):

“The key secondary endpoints of progression-free survival, objective response rate, and alive and progression-free at 6 months were all demonstrated with statistical significance, showing improvement in favor of selumetinib in combination with docetaxel versus docetaxel alone.”

Indeed, at the recent AACR and ASCO meetings, there was also some encouraging early signs from Genentech’s PI3K inhibitor, GDC-0941, as a single targeted agent with chemotherapy in NSCLC (a very small early trial), albeit not KRAS specific, but defined more broadly by squamous and non-squamous histology. Thus, all is not lost with the MEK agent yet – if we combined MEK and PI3K inhibitors in NSCLC patients previously treated with chemotherapy, we might have a better chance of succeeding and shutting down the pathway, based on evidence offered from Wee et al’s preclinical research:

“At the molecular level, the dual inhibition of both pathways seems to be required for complete inhibition of the downstream mammalian target of rapamycin effector pathway and results in the induction of cell death.”

As a result, they went onto to suggest a logical treatment approach:

“Our study provides molecular insights that help explain the heterogeneous response of KRAS mutant cancers to MEK pathway inhibition and presents a strong rationale for the clinical testing of combination MEK and PI3K targeted therapies.”

Of course, clinical trials like this always progress incrementally, such that we test a MEK or a PI3K inhibitor alone to determine safety and efficacy activity, then perhaps in combination, which requires another phase I dose finding study to determine the ideal dosages and whether they are too toxic or not combined.

So while either single agent targeted therapy with chemotherapy in and of itself is not a win, there are signs that combining the two may be more appropriate. I would still want to know what is driving the KRAS activity though, given MEK and PI3K are downstream of it. It is entirely possible that a third agent would be needed to shut down the pathway more completely in that patient subset.

At ECCO, De Bono (Royal Marsden) discussed the combination of AstraZeneca’s MEK inhibitor (AZD6244) and Merck’s AKT inhibitor (MK-2206) in RAS mutant colorectal (CRC) and lung (NSCLC) cancers. The results here were not a big win in the former, with 8/15 patients showing no antitumour activity to date.

There are several things we can conclude from the initial data:

  • If we have the right combination for the right target in the right patient subset, then the therapeutic index of the agents is lacking and we need better drugs
  • Are the targets (AKT and MEK) critical?
  • Is something else driving the KRAS activity (see below)*?
  • Are we shutting down the adaptive resistance pathways (escape routes?)
  • Which patient subsets are most likely to respond and how do we best characterise them (ie need more biomarker data)?

And so on… there are always more questions than answers sometimes.

    * Note: This situation could well be similar to BRAF in malignant melanoma, where it is the V600E mutation that is driving the BRAF activity, thus specifically targeting ithe mutation rather than the kinase will have a greater clinical effect than targeting BRAF broadly. In this case, if we really believe KRAS is critical to the lung or colorectal tumour’s survival, then we need to figure out what is driving it before progress is made. Frank McCormick’s elegantly simple wac-a-mole concept for pathway inhibition is very apt here!

No doubt we will see more detailed data and an update soon, perhaps even at the forthcoming AACR Molecular Targets meeting next month.

References:

ResearchBlogging.orgWee, S., Jagani, Z., Xiang, K., Loo, A., Dorsch, M., Yao, Y., Sellers, W., Lengauer, C., & Stegmeier, F. (2009). PI3K Pathway Activation Mediates Resistance to MEK Inhibitors in KRAS Mutant Cancers Cancer Research, 69 (10), 4286-4293 DOI: 10.1158/0008-5472.CAN-08-4765

4 Comments

On mTOR and Tuberosis Sclerosis Complex (TSC)

By on July 26, 2011

It’s that time of year in the dog days of summer when many people in the industry are either incredibly busy, heads down, rolling out new things for the third quarter or else it’s a pleasant lull between the strategic and tactical phases and a good time to catch your breath.  Here in the Icarus office, we’re busy creating and writing a new series of syndicated reports in a variety of different tumour types and pathways.  I have hundreds of snippets and notes saved electronically from various cancer meetings this year, making it a great opportunity to collate and process them into broader insights. If you have any particular needs in this area, now is a good time to let us know, so do email me and your wishes may get added to the list.

Last week I was in Boston and happened by chance to walk past the Whitehead Institute. This reminded me that I had David Sabatini’s new mTOR paper in Science queued up to blog about on Pharma Strategy Blog.

The mTOR pathway is highly complex and consists of a huge network of interwined proteins and kinases:

Source: wikipedia

Hsu et al., (2011) described what they found from defining the mTOR-regulated phosphoproteome using quantitative mass spectrometry and protein libraries to build a complete picture:

“The adaptor protein Grb10 was identified as an mTORC1 substrate that mediates the inhibition of phosphoinositide 3-kinase typical of cells lacking tuberous sclerosis complex 2 (TSC2), a tumor suppressor and negative regulator of mTORC1.

Our work clarifies how mTORC1 inhibits growth factor signaling and opens new areas of investigation in mTOR biology.”

We know, for example, that mTORC1 inhibits PI3K-Akt signaling, but the precise molecular connections involved are poorly understood.  S6K1 phophosphorylation, which destabilises the insulin receptor substrate 1 (IRS1), is one mechanism known to be involved.  Hsu et al., demonstrated that other mechanisms are also critical:

“mTORC1 inhibits and destabilizes IRS1 and simultaneously activates and stabilizes Grb10.”

They went to separate the effects of acute and chronic stimulation of mTOR:

“Whereas acute mTORC1 inhibition leads to dephosphorylation of IRS1 and Grb10, chronic mTORC1 inhibition leads to changes in the abundance of IRS and Grb10 proteins, which are likely the most important effects of mTOR inhibitors to consider in their clinical use.”

This important article is particularly relevant because not long after the publication, Novartis announced positive data with their mTOR inhibitor, everolimus (Afinitor) in patients with tuberous sclerosis complex (TSC).

TSC is a genetic disorder that affects affects approximately 1-2 million people worldwide and is associated with a variety of resulting disorders including seizures, swelling in the brain, developmental delays and skin lesions. It can also cause non-cancerous tumours to form and can affect many different parts of the body such as the brain and kidney, for example.

The rationale behind such as study was described in the Novartis press release:

“Tuberous sclerosis complex is caused by defects in the TSC1 and/or TSC2 genes. When these genes are defective, mTOR activity is increased, which can cause uncontrolled tumor cell growth and proliferation, blood vessel growth and altered cellular metabolism, leading to the formation of non-cancerous tumors throughout the body, including the brain.”

In other words, giving an mTOR inhibitor such as everolimus may help by reduce cell proliferation, blood vessel growth and glucose uptake associated with the TSC defect.

In patients with brain lesions, surgery is usually considered the only viable option, so a study showing a 35% response rate (50% reduction or more) in the SEGA lesions, is a positive step forward.  The new data was from a phase III trial (n=117) and appears to support the initial positive phase II study, so it will likely lead to a registration filing in this indication for everolimus.

 

References:

ResearchBlogging.orgHsu, P., Kang, S., Rameseder, J., Zhang, Y., Ottina, K., Lim, D., Peterson, T., Choi, Y., Gray, N., Yaffe, M., Marto, J., & Sabatini, D. (2011).  The mTOR-Regulated Phosphoproteome Reveals a Mechanism of mTORC1-Mediated Inhibition of Growth Factor Signaling. Science, 332 (6035), 1317-1322 DOI: 10.1126/science.1199498

New mechanisms of resistance to MET inhibition

By on April 12, 2011

A very apt quote from Jeff Engelman’s group caught my eye this week:

“Unfortunately, cancers invariably develop resistance, and overcoming or preventing resistance will ultimately be key to unleashing their full therapeutic potential.”

MET is the receptor tyrosine kinase for hepatocyte growth factors (HGF) and inhibition has been implicated in metastases and migration of cancer cells (Rong et al., (1994), Takayama et al., (1997)), but more recently, it has also been observed that some tumour types have MET oncogenic addiction, including gastric cancer (Smolen et al., 2006).

MET Inhibition - source: Ma et al., (2007) BJC

Qi et al., (2011) went on to explain how they were looking at strategies for overcoming resistance to MET inhibitors, using PHA-665752 and PF-2341066, as an example in highly sensitive gastric cell lines.  They investigated the possibilities in vivo and in vitro. The results, however, were unexpected:

“To our surprise, we observed at least two mechanisms of resistance that arose simultaneously.  Both resulted in maintenance of downstream PI3K (phosphoinositide 3-kinase)-AKT and MEK (MAP/ERK kinase)-ERK signaling in the presence of inhibitor.”

Many of you will be aware of activation loops from other kinases, such as imatinib (Gleevec) in CML (T315I) and GIST (D842V) or erlotinib (Tarceva) in lung cancer (T790M), and adaptive pathways e.g. with BRAF inhibitors such as PLX4032 (vemurafenib) in melanoma, so this phenomenon is not uncommon.

With the MET inhibitors tested in the current research, the group found:

  1. A mutation in the MET activation loop, Y1230
  2. Activation of the epidermal growth factor receptor (EGFR) pathway due to increased expression of transforming growth factor alpha (TGFa)

What do these results mean?

The data suggests that combining MET and EGFR inhibitors in gastric cancer may be a viable therapeutic strategy, but consideration must also be given to approaches that inhibit Y1230 mutant MET as well, in order to shut off the escape routes.

It is hard to argue with the authors conclusion that:

“These results also underscore the notion that a single cancer can simultaneously develop resistance induced by several mechanisms and highlight the daunting challenges associated with preventing or overcoming resistance.”

Given the positive results seen with trastuzumab (Herceptin) in patients with HER2-positive gastric cancer, part of me is also wondering what incremental value there would be efficacy-wise, if MET and EGFR inhibitors were used in combination with trastuzumab?  We know that the blocking the driver mutation, the adaptive pathway and the ligand is important.  Some further preclinical research in this area may shed light on the matter.

References:

ResearchBlogging.orgQi, J., McTigue, M., Rogers, A., Lifshits, E., Christensen, J., Janne, P., & Engelman, J. (2011). Multiple Mutations and Bypass Mechanisms Can Contribute to Development of Acquired Resistance to MET Inhibitors Cancer Research, 71 (3), 1081-1091 DOI: 10.1158/0008-5472.CAN-10-1623

Ma, P., Tretiakova, M., Nallasura, V., Jagadeeswaran, R., Husain, A., & Salgia, R. (2007). Downstream signalling and specific inhibition of c-MET/HGF pathway in small cell lung cancer: implications for tumour invasion British Journal of Cancer, 97 (3), 368-377 DOI: 10.1038/sj.bjc.6603884

Rong S, Segal S, Anver M, Resau JH, & Vande Woude GF (1994). Invasiveness and metastasis of NIH 3T3 cells induced by Met-hepatocyte growth factor/scatter factor autocrine stimulation. Proceedings of the National Academy of Sciences of the United States of America, 91 (11), 4731-5 PMID: 8197126

Takayama H, LaRochelle WJ, Sharp R, Otsuka T, Kriebel P, Anver M, Aaronson SA, & Merlino G (1997). Diverse tumorigenesis associated with aberrant development in mice overexpressing hepatocyte growth factor/scatter factor. Proceedings of the National Academy of Sciences of the United States of America, 94 (2), 701-6 PMID: 9012848

Smolen GA, Sordella R, Muir B, Mohapatra G, Barmettler A, Archibald H, Kim WJ, Okimoto RA, Bell DW, Sgroi DC, Christensen JG, Settleman J, & Haber DA (2006). Amplification of MET may identify a subset of cancers with extreme sensitivity to the selective tyrosine kinase inhibitor PHA-665752. Proceedings of the National Academy of Sciences of the United States of America, 103 (7), 2316-21 PMID: 16461907

error: Content is protected !!