Intended for healthcare professionals

Letters

Tumour markers in malignancies

BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7257.379/a (Published 05 August 2000) Cite this as: BMJ 2000;321:379

Two isoforms of oestrogen receptor are now known to exist

  1. Carlo Palmieri (c.palmieri{at}ic.ac.uk), CRC clinical research fellow,
  2. Sam Fishpool, medical student,
  3. R C Coombes, professor of medical oncology
  1. Cancer Cell Biology Group, Cancer Research Campaign Laboratories, Imperial College School of Medicine—Hammersmith Campus, London W12 0NN
  2. St George's Hospital Medical School, Division of Oncology, Department of Cellular and Molecular Sciences, London SW17 0RE
  3. Canterbury Health, Department of Medicine, Christchurch Hospital, PB 4710, Christchurch, New Zealand
  4. Department of Clinical Genetics, Karolinska Hospital, Stockholm, Sweden
  5. Department of South Stockholm Oncology, Huddinge University Hospital, Sweden

    EDITOR—Lindblom and Liljegren comment about oestrogen receptor in their clinical review of tumour markers in malignancies.1 They failed, however, to acknowledge that two isoforms of the receptor (ERa and ERb) are now known to exist and to distinguish between the two. This is important as it is now incorrect both scientifically and clinically to talk solely about ER as if it were one entity.

    The oestrogen receptor that they referred to is the classical oestrogen receptor or ERa2; as noted in their table, it predicts response to endocrine treatment in the adjuvant setting and correlates with a better prognosis. Not noted in their table is that ERa immunohistochemistry is used as a diagnostic investigation in patients with metastases when there is a primary of unknown origin. Also not noted is that ERa expression does not guarantee response to endocrine treatment, with 30-40% of such tumours failing to respond.3

    Oestrogen receptor β, or ERb, is the more recently discovered isoform; ERa and ERb represent two gene products with distinct biological roles and ligand binding specificity.4 With reference to the breast the expression of ERb, its role in the normal and malignant breast, the interactions between it and ERa, and its use as a tumour marker are currently being investigated. Recent immunohistochemical study of ERb in breast cancer, however, has shown it to correlate with ERa positivity, low grade, and negative axillary lymph node status—that is, good prognostic factors.5 The independent predictive value of ERb has yet to be established.

    The presence of a further oestrogen receptor has opened up new avenues of research; these will lead to a clearer understanding of hormonally dependent breast cancer and more precise methods of predicting response to hormonal treatment. In addition, more effective hormonal treatment based on receptor specificity might be developed.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.

    CA19.9 is useful in several cancers …

    1. A G Dalgleish, professor of oncology
    1. Cancer Cell Biology Group, Cancer Research Campaign Laboratories, Imperial College School of Medicine—Hammersmith Campus, London W12 0NN
    2. St George's Hospital Medical School, Division of Oncology, Department of Cellular and Molecular Sciences, London SW17 0RE
    3. Canterbury Health, Department of Medicine, Christchurch Hospital, PB 4710, Christchurch, New Zealand
    4. Department of Clinical Genetics, Karolinska Hospital, Stockholm, Sweden
    5. Department of South Stockholm Oncology, Huddinge University Hospital, Sweden

      EDITOR—I was surprised that Lindblom and Liljegren mentioned CA19.9 almost as an aside as being at the experimental stage in ovarian cancer.1 This marker is associated with other adenocarcinomas, particularly pancreatic cancer; when used in conjunction with knowledge of the patient's clinical state it can be an excellent guide to diagnosis and response to treatment. It has been routinely used in the south west of London for some time, and a survey of PubMed shows that many papers have reported its usefulness in the management of pancreatic cancer.

      Another tumour marker, which the authors have omitted altogether, is lactic dehydrogenase; this is one of the most useful indicators of relapse melanoma as well as other malignancies. I am surprised that these omissions escaped the peer review process.

      References

      1. 1.

      … as are monoclonal immunoglobulin and β human chorionic gonadotrophin

      1. B M Colls
      1. Cancer Cell Biology Group, Cancer Research Campaign Laboratories, Imperial College School of Medicine—Hammersmith Campus, London W12 0NN
      2. St George's Hospital Medical School, Division of Oncology, Department of Cellular and Molecular Sciences, London SW17 0RE
      3. Canterbury Health, Department of Medicine, Christchurch Hospital, PB 4710, Christchurch, New Zealand
      4. Department of Clinical Genetics, Karolinska Hospital, Stockholm, Sweden
      5. Department of South Stockholm Oncology, Huddinge University Hospital, Sweden

        EDITOR—Lindblom and Liljegren's review of tumour markers in malignancies is particularly interesting to those interested in the genetic and molecular biological aspects of this subject.1 Concentration on these topics, however, has resulted in the authors failing to emphasise the importance of the tumour markers that have been in clinical use for many years.

        No mention is made of monoclonal immunoglobulin, which is certainly the oldest tumour marker known and is probably the most extensively used. Indeed, the presence of monoclonal immunoglobulin in blood or urine is one of the prerequisites in diagnosing myeloma. Its measurement is also used in staging and prognosis, and changes after treatment are helpful in management.2 Thus it is almost the perfect tumour marker in terms of diagnosis, staging, and management and certainly warrants consideration in such a review.

        Scant regard is paid to the use of β human chorionic gonadotrophin and α fetoprotein in the diagnosis, staging, and management of testicular and ovarian germ cell tumours. Lindblom and Liljegren note that the management of testicular cancer has improved considerably, and this is in large part due to the study of these tumour markers. In particular, the successful management of stage 1 non-seminomatous germ cell testicular tumour depends heavily on the presence or absence or changing levels of these tumour markers.3

        Finally, β human chorionic gonadotrophin is useful in the management of molar pregnancies or subsequent choriocarcinoma, or both.4 Choriocarcinoma as a cause of death now rarely occurs because of the measurement of β human chorionic gonadotrophin after molar pregnancy. Surely this excellent tumour marker warrants at least a mention in such a review.

        References

        1. 1.
        2. 2.
        3. 3.
        4. 4.

        Authors' reply

        1. Annika Lindblom, clinical geneticist (Annika.lindblom{at}cmm.ki.se),
        2. Annelie Liljegren, oncologist
        1. Cancer Cell Biology Group, Cancer Research Campaign Laboratories, Imperial College School of Medicine—Hammersmith Campus, London W12 0NN
        2. St George's Hospital Medical School, Division of Oncology, Department of Cellular and Molecular Sciences, London SW17 0RE
        3. Canterbury Health, Department of Medicine, Christchurch Hospital, PB 4710, Christchurch, New Zealand
        4. Department of Clinical Genetics, Karolinska Hospital, Stockholm, Sweden
        5. Department of South Stockholm Oncology, Huddinge University Hospital, Sweden

          EDITOR—Palmieri et al emphasise the role of the recently identified oestrogen receptor β. This may indeed become a useful tumour marker in the future. For our review, however, we were asked to include only tumour markers known to be of common clinical use. That is why we did not discuss several new and potentially interesting markers.

          Dalgleish points out that in many hospitals the marker CA19.9 is commonly used clinically, not just experimentally. The use of many markers varies, and we tried to select those most commonly used worldwide and point them out as being clinically used; that is why we said that CA125 was in common clinical use and CA19.9 was still experimental. The marker lactate dehydrogenase is commonly used in follow up in lymphoma (as well as in other malignancies). This marker was included in the original version of our article, but we eventually excluded it because it is an unspecific marker of cell decomposition (destruction) and does not fit the criteria we were told to use for a tumour marker. Besides, it is used primarily as a prognostic marker, and we removed all prognostic markers on request from the reviewer.

          Colls points out the usefulness of the marker β human chorionic gonadotrophin in the management of molar pregnancies and subsequent choriocarcinoma. This marker is definitely of use, but, perhaps erroneously, we omitted it and other markers used in very rare diseases.

          Finally, we did not include the commonly used marker monoclonal immunoglobulin in myeloma in our review. This we regret; it fits all the criteria we finally used in selecting the markers to be included.

          We emphasise that our article was meant to be a broad overview and to include selected markers used in malignancies. We realise that other doctors find many more markers to be important. We ourselves were strongly influenced in the selection of markers by the reviewer's guidance; this further shows that the subject is not clear cut and that there are almost as many opinions as there are doctors.

          View Abstract

          Log in

          Log in through your institution

          Subscribe

          * For online subscription