Cancer can occur in any organ of the human body. Individual cells begin to reproduce in greater numbers than normal, the surrounding tissue cannot restrain the growth, and then they escape and invade other tissues or organs. Cancer commonly occurs as a result of genetic defects.

Pancreatic cancer has one of the worst prognoses of all human malignancies. It is an aggressive disease and the overall 5-year survival rate is less than 1%. Between 75% and 85% of patients present with inoperable tumors. Unfortunately, conventional oncological approaches such as radiotherapy and chemotherapy have failed to improve the prognosis in a relevant manner.

“Pancreatic cancer has one of the worst prognoses of all human malignancies.”

In the past two decades many laboratories have concentrated on understanding the molecular alterations that are present in pancreatic cancer. That’s my area of research, and our goal is to improve diagnosis and prognosis.

I did my medical training at the University of Tübingen in Germany and then at the University of Munich. For the last part of this medical training I went to Switzerland to the University of Bern, where I met the research group under Professor Helmut Friess and Professor Markus Büchler. They were mainly interested in pancreatic surgery and research, and it was through them that I entered the field. I still work with both of them.

My postdoctoral training was at the University of California, Irvine (Professor Murray Korc), for two years, and that’s where I really started research on the molecular biology of pancreatic cancer. After those two years I returned to Bern, stayed for three more years, and then we all moved here, to Heidelberg.

We’re within the surgical department of the University of Heidelberg, which is one of the leading centers worldwide, with about 400 pancreatic operations a year. Importantly, we have established a very strong center for molecular biology and translational research, and next door we have the German Cancer Research Centre, with whom we have several collaborations. The infrastructure for our research is first rate.

The focus today has shifted somewhat from the period seven or eight years ago that resulted in the highly cited papers we’ll discuss in a moment; back then we did not know a lot, and were just cherry-picking. Now we have a different approach, with high throughput analysis, particularly high-volume micro-array analysis. Today we can fish out from thousands of genes the few that might be relevant to pancreatic cancer.

If we are going to beat this disease we really need to find the genes or the genetic alterations that are important for pancreatic cancer, with the ultimate goal to find some new markers for the disease and therapeutic targets. This we have been doing for some years, and we have now identified several candidate genes. One part of our research is to look at these different genes in more depth and see if there are candidates for therapeutic intervention.

  Your 1999 Oncogene paper ("The TGF-beta signaling inhibitor Smad7 enhances tumorigenicity in pancreatic cancer," 18[39]: 5363-72, 23 September 1999) and your 1999 Biochemical and Biophysical Research Communications paper ("Smad6 suppresses TGF-beta-induced growth inhibition in COLO-357 pancreatic cancer cells and is overexpressed in pancreatic cancer," 255[2]: 268-73, 16 February 1999) are about the inhibitory Smads, 6 and 7. What is the importance of these signaling molecules in pancreatic cancer?

Let me first give the background to this interesting topic. Before we did this work we already understood that there were certain growth factors like the transforming growth factor (TGF) beta family that we knew were over-expressed in pancreatic cancer. The tumors produce a lot of these factors, and at the same time it was known that patients who produce a lot have a worse survival. We also knew that these factors were suppressing the growth of tumor cells. So it was quite a puzzle: you have a factor that suppresses tumor growth and yet at the same time the tumor is producing it!

The puzzle was resolved in 1996 by Hahn and colleagues (Science 271: 350-353) with the breakthrough discovery of a mutation in the central aspect of the TGF-beta signaling pathway (Smad4/DPC4). But the mutation was only observed in about 50% of the cancers, so it was still not clear what happened in the other tumors. So we looked at recently discovered members of this signaling pathway: the Smad6 and Smad7 molecules. These can act as inhibitors of the TGF-beta signaling pathway.

What we showed was that these inhibitory Smads are actually also over-expressed in pancreatic cancer; that provided another way for the tumors to become resistant to the growth-suppressive effects of TGF-beta.

We were among the first to show that inhibitory Smads are important in pancreatic cancer as well as in cancer in general. Our achievement was showing how the tumor cells could become resistant to the growth-suppressive effects of TGF-betas.

These papers are highly cited because they showed how this signaling pathway is altered in pancreatic cancer; it’s not restricted just to the case of the pancreas but is important for other neoplastic and non-neoplastic diseases. We have learned over the years that this TGF-beta signaling pathway is one key factor in pancreatic carcinogenesis.

This paper is related to the previous one. The heparan sulfate proteoglycans are very complex molecules, which constitute prominent components of basement membranes and extracellular matrix (ECM). What we knew was that there are certain enzymes that are able to modify these molecules by cutting their side chains, thus facilitating disassembly of the ECM and enhancing cell invasion.

One of these degrading enzymes is heparanase. By the time we were doing the research on the previous paper the protein had been identified but not cloned. So when the first reports appeared of the cloning of this enzyme we were interested to see whether this was also important in pancreatic cancer.

We showed that there is increased expression of heparanase in pancreatic cancer. It can change certain molecules and this enables the cells to invade the microenvironment of the tumor. That means the tumor spreads more, leading to early metastasis. This has clinical relevance because patients with tumors that express a lot of this enzyme have a worse clinical outcome.

These are different: Id1 and Id2 are transcription factors. We became interested because at the time if you looked at pancreatic cancer research you could see that there were a fairly large number of alterations that had already been identified. The question remained as to the key alterations that drive all the other changes. To find the key alterations you have to search for transcription factors. We showed that the helix-loop-helix proteins Id1 and Id2 were present in the tumor cells.

This is a highly cited paper because it was one of the first showing that these transcription factors have an important role in pancreatic cancer and also in cancer generally. The Id proteins are versatile molecules that have a prominent role in cancer progression.

Jörg Kleeff, MD
Department of General Surgery
University of Heidelberg
Heidelberg, Germany