Ultrasound can Predict Tumor Burden and Survival in Melanoma Patients, Sparing Unnecessary Surgery

Ultrasound can predict tumor burden and survival in melanoma patients. Researchers have shown for the first time that patterns of ultrasound signals can be used to identify whether or not cancer has started to spread in melanoma patients, and to what extent. The discovery enables doctors to decide on how much surgery, if any, is required and to predict the patient’s probable survival.

Dr Christiane Voit told Europe’s largest cancer congress, the joint 15th European CanCer Organisation (ECCO) and 34th European Society for Medical Oncology (ESMO) Multidisciplinary Congress, September 20 – 24 in Berlin, Germany: “We have identified two ultrasound patterns of lymph node metastasis in melanoma patients which can identify correctly any amount of tumor cells in the sentinel lymph nodes in 75-90% of cases before proceeding to surgery on the sentinel lymph nodes.”

Voit, who is a dermatologist and head of the diagnostic unit at the Skin Cancer Centre at Charité – Universitätsmedizin Berlin, the Medical University of Berlin, said that although her research needs to be confirmed in multi-centre, randomized clinical trials, it had the potential to spare patients unnecessary surgery, especially if it was combined with ultrasound-guided fine needle biopsy of lymph nodes rather than conventional surgery.

Since 2001 Dr Voit and her colleagues in Germany and The Netherlands have included 850 melanoma patients in a prospective study to investigate the use of ultrasound in diagnosis and treatment planning. They have already demonstrated that ultrasound-guided fine needle biopsy of sentinel nodes before conventional sentinel node surgery can identify up to 65% of patients in whom the cancer has started to spread. The study presented today shows how far ultrasound patterns correlate with disease progression, tumor burden, survival and prognosis in the first 400 of these patients with stage I/II melanoma and with the longest follow-up.

Before having sentinel node surgery the patients were investigated using ultrasound, and these results were checked against the results of the subsequent surgery. The researchers found that two ultrasound patterns together could correctly identify the amount of cancer cells in the lymph nodes in 80% of cases.

A balloon shape ultrasound pattern with or without loss of central echoes (where the lymph node has lost central echoes or still has some residual central echoes, but these are wandering toward the rim, giving an asymmetrical shape to the centre) was an indicator in up to 83% of cases of a large amount of cancer cells in the sentinel node. “This ultrasound pattern was a late sign, only occurring in cases of advanced metastasis,” said Voit.

A pattern of peripheral perfusion (where small blood vessels start to surround the lymph node) was an early sign of a small number of cancer cells present. “The early signs are signs of first disruption of the normal lymph node architecture by an early stage metastasis. The most important one is peripheral perfusion, which shows angiogenesis (the formation of new blood vessels) is occurring,” she explained.

The researchers found that these two ultrasound patterns could predict overall survival. Estimates for overall survival after five years for patients with stage I/II is between 50-90% depending on the state of the tumour. Dr Voit found that 93% of patients with neither of these ultrasound patterns, 87% of patients with peripheral perfusion, and 56% of patients with balloon shapes with or without loss of central echoes, survived for at least five years; survival without cancer spreading to other parts of the body was 74%, 60% and 26% respectively.

Dr Voit said: “For the first time we have established that ultrasound patterns can be used as criteria for diagnosing disease progression and tumor burden. Balloon shaped lymph nodes with or without loss of central echoes and peripheral perfusion are independent prognostic factors for survival.”

Discovering if cancer has spread to the lymph nodes is the most important factor influencing the prognosis and treatment of melanoma patients. Doctors usually cut out one or two key lymph nodes, called sentinel nodes, and use these as an indicator of whether or not the cancer has spread to the other lymph nodes. If the sentinel node is free of cancer, patients don’t need to have more extensive lymph node removal.

However, only 20% of patients who have a sentinel node biopsy have cancer that has spread there, and so the operation, which can be accompanied by side effects such as chronic swelling and seroma, is unnecessary for 80% of patients. Using ultrasound first to detect the presence or not of sentinel node metastases could be a non-invasive way of limiting the numbers of patients who require subsequent surgery or simply watchful follow-up care.

For more information:

• Abstract no: 9303. Melanoma session, Wednesday 14.45-17.00 hrs CEST (Hall 7)

Also read these PubMed abstracts:

• Voit C, Kron M, Schäfer G, Schoengen A, Audring H, Lukowsky A, Schwürzer-Voit M, Sterry W, Winter H, Rademaker J. Ultrasound-guided fine needle aspiration cytology prior to sentinel lymph node biopsy in melanoma patients. Ann Surg Oncol. 2006 Dec;13(12):1682-9.
• Voit CA, Schäfer-Hesterberg G, Kron M, van Akkooi AC, Rademaker J, Lukowsky A, Schoengen A, Schwürzer-Voit M, Sterry W, Krause M, Röwert-Huber J, Eggermont AM.
  Impact of molecular staging methods in primary melanoma: reverse-transcriptase polymerase chain reaction (RT-PCR) of ultrasound-guided aspirate of the sentinel node does not improve diagnostic accuracy, but RT-PCR of peripheral blood does predict survival. J Clin Oncol. 2008 Dec 10;26(35):5742-7. Epub 2008 Nov 3.

Images courtesy American Society of Clinical oncology (ASCO).

Identification of Highly Radiosensitive Patients May Lead to Side Effect-free Personalized Radiotherapy

An international group of scientists has taken the first step on the road to targeting radiotherapy dosage to individual patients by means of their genetic characteristics, a radiation oncologist told Europe’s largest cancer congress, ECCO 15 – ESMO 34, the 15th congress of the European CanCer Organization and the 34th congress of the European Society for Medical Oncology, in Berlin today.

Professor Dirk de Ruysscher, from Maastricht University Medical Centre, Maastricht, The Netherlands, said that his team’s work might provide the basis for personalized radiotherapy in which, with a simple blood test, doctors may be able to select the optimal radiation dose for a particular patient.

The team of scientists from The Netherlands, Belgium, Germany, and Canada studied a group of patients with hypersensitivity to radiation therapy, drawn from the largest world-wide database available – the European Union-funded Genetic pathways for the prediction of the effect of irradiation (GENEPI) study, which integrates biological material with patient data and treatment specifications. The database included information from more than 8000 European patients.

“Part of this project is the establishment of a sub-database in which very rare patient characteristics are brought together with the hypothesis that their genetic traits will enable the characterization of molecular pathways related to radio-sensitivity,” explained Professor de Ruysscher. “A major problem for radiation oncologists at present is that we are bound by the need to avoid damage to normal tissues. This means that the dose of radiation generally used is governed by the response of the most radiosensitive patients, and this may lead to many patients receiving lower than optimal doses, hence affecting the ability to deliver a higher dose that may result in better local tumor control.”

A tissue bank including skin fibroblasts (the structural framework of skin cells), whole blood, lymphocytes (white blood cells involved in the immune system), plasma, and lymphoblastic (immature lymphocyte) cell lines from patients who were known to be hypersensitive to radiation was established from patients in Europe and Canada.

When compared with a control group, also drawn from the GENEPI study, the hypersensitive patients showed either severe side effects occurring at very low radiation levels, or severe side effects lasting for more than four weeks after the end of radiotherapy and/or requiring surgery, or severe late side effects occurring or persisting more than 90 days after the end of radiotherapy.

The scientists identified 33 such patients, 10 males and 23 females, of whom 11 (two males and nine females) ultimately proved to be really hypersensitive to radiation, underlining the rarity of this condition. Their mean age was 61.6 ± 8.5 years (range 49 – 74). One patient had non-small cell lung cancer, six breast cancer, two head and neck cancer and one lymphoma. The radiation doses, the overall treatment times, and the follow-up times all fell within the usual parameters.

The mean radiation dose to the tumor was 45.3 ± 18.3 Gy (range 8 – 66), delivered in a mean of 21.5 ± 10.5 fractions (range 1 – 33), in a mean overall treatment time of 31.4 ± 17.6 days (range 1 – 57). The mean follow-up time after radiotherapy was 1658 ± 1048 days (range 84 – 3752).

“The severe side effects included acute skin reactions, extreme skin thickening or fibrosis, lung tissue inflammation and blindness due to optical nerve damage,” said Professor de Ruysscher. “Although radiotherapy is a highly effective way of treating cancer, it is important that we are able to identify the patients who will react badly to it and adjust their dosage accordingly.”

Radiotherapy works by causing DNA damage in cells in a particular area so that they destroy themselves. Because cancer cells reproduce more and are undifferentiated (lacking the ability to become a more specialized cell type), they are less able to repair the damage caused by radiotherapy than are differentiated, normal cells which can usually repair themselves. However, some of the normal cells surrounding the treatment site may also be damaged during radiotherapy, and it is this damage that leads to side effects.

Scientists already know that different types of tumors respond differently to radiotherapy; highly radiosensitive cancer cells such as leukemia can be killed by quite low radiation doses, whereas melanomas need such a high dose that it would be unsafe to use radiation therapy in this case. The finding that individuals, as well as tumors, react differently will enable doctors in the future to target doses even more carefully, taking into account not just the radiosensitivity of the tumor type but also the potential reaction of the particular patient to treatment.

“We hope that the EU will fund a successor project to elucidate genetic pathways in combination with other patient data so that we can make predictive models that can be implemented in standard clinical practice,” said Professor de Ruysscher. “We believe that, if we can understand what it is going on at a molecular level, we may be able to develop a blood test that will allow us to know precisely how an individual patient will react to radiotherapy, and to target the dose accordingly. Such personalized treatment will be a major advance, allowing us to minimize both radiotherapy doses and unpleasant side effects, while treating the tumor in the most effective way possible. Perhaps even more importantly, it will enable us to give higher doses to many patients and hence improve control of their tumors.”

For more information

• Abstract no: 2007, Radiotherapy and Radiobiology session, Thursday 10.45 hrs CEST (Hall 15.3)
• Genetic Pathways for the Prediction of the Effects of Ionising Radiation: Low Dose Radiosensitivity and Risk to Normal Tissue after Radiotherapy
• GENEtic pathways for the Prediction of the effect of Irradiation-European normal an tumour tissue bank and data base