DNA Ploidy and Cell Cycle Analysis in Cancer Diagnosis and Prognosis

June 1, 1996

pplication of the techniques of flow cytometry and image analysis to quantitation of DNA and estimation of events in the cell cycle in human tumors has achieved considerable popularity as a laboratory procedure but so far has failed to be of practical clinical value. As brilliantly summarized by Dr. Ross, retrospective studies have repeatedly shown abnormal DNA values to be of prognostic significance in several organ systems, among them, tumors of the urothelium [1] and prostate [2] and, perhaps to a lesser extent, mammary carcinomas [3].

Application of the techniques of flow cytometry and image analysisto quantitation of DNA and estimation of events in the cell cyclein human tumors has achieved considerable popularity as a laboratoryprocedure but so far has failed to be of practical clinical value.As brilliantly summarized by Dr. Ross, retrospective studies haverepeatedly shown abnormal DNA values to be of prognostic significancein several organ systems, among them, tumors of the urothelium[1] and prostate [2] and, perhaps to a lesser extent, mammarycarcinomas [3].

Unfortunately, few prospective studies with long follow-up havedocumented in a persuasive fashion the impact of these measurementson treatment. At best, tumor ploidy patterns are considered asancillary information in treatment protocols, along with a hostof other factors. Although DNA tests are often requested by interestedoncologists and surgeons, and the results of these tests are providedby competent laboratories, there is limited evidence that tumorploidy is an important factor in the management of malignant tumors.The laboratory reports are simply filed away as a part of theclinical records.

It is quite possible that information on the significance of DNAstudies has not been appropriately disseminated, and therefore,is inaccessible to practicing medical or surgical oncologists,let alone other practitioners. For example, it is nearly routineto measure DNA in samples of mammary carcinoma. Is this informationstudied? Is it understood? Is treatment modified as a consequenceof the DNA studies? More importantly, perhaps, is the informationshared with the patient?.

DNA Analysis Has Great Clinical Significance in Some Tumors

Ironically, in some areas, DNA ploidy measurement could be ofgreat clinical significance. For example, the treatment of prostaticcarcinoma could be significantly influenced by the use of DNAdata. Several independent sources have provided reliable evidencethat diploid prostatic cancers have a relatively indolent courseand progress very slowly [4-6]. It is quite likely that patientswith such tumors, particularly those of low stage, could be observedor treated conservatively with hormonal therapy without any significantloss in long-term survival. In fact, the quality of life of thepatients not treated by radical extirpation or massive radiotherapyof the prostate would very likely be much superior to that oftreated patients.

There also is evidence that DNA ploidy measurements may have amajor impact on the behavior of high-stage and even metastaticprostatic cancers. Tumors in the diploid range and some tumorsin the tetraploid range progress slowly and show a better survivalthan aneuploid tumors [7,8]. On the other hand, aneuploid prostaticcarcinomas are likely to progress rapidly, and therefore, requirerapid, aggressive treatment.

Despite these findings, there is no evidence that DNA analysishas had a major impact on the ever-increasing frequency of radicalprostatectomies [9]. One wonders why this is so, particularlyin the absence of hard data on clinician's attitudes. It is likelythat physicians harbor significant lingering doubts as to thereliability and clinical value of the test, particularly in theabsence of guidelines from major medical centers.

What DNA Ploidy Measurement Does and Doesn't Tell Us

What does the information from DNA analysis tell us? MeasuringDNA and cell cycle events in a tissue or cell sample providesonly coarse information on the genetic make-up of the tumor. Theinformation is not of diagnostic value because, as Dr. Ross emphasizes,benign tumors can be aneuploid and malignant tumors diploid [10].It does tell us that, in some malignant tumors, the changes inthe cell genome have been slight or none (diploid tumors) or marked(aneuploid tumors). Interestingly, diploid solid malignant tumorsoccurring in adults that retain a complete or nearly completegenome have, on the whole, a less threatening behavior than aneuploidtumors. Although the proof is lacking, it appears likely that,in many but not all diploid tumors, the mechanisms regulatingcell proliferation are relatively intact. Oddly enough, as Dr.Ross stresses, small cell diploid tumors of childhood respondpoorly to therapy, whereas aneuploid tumors do much better.

There are limits to what we can infer from DNA ploidy measurement,however. For example, DNA analysis does not tell us anything aboutpossible genetic changes, such as activation of tumor-promotinggenes (oncogenes) or mutation of tumor-inhibitory genes (anti-oncogenes).Expensive, time-consuming molecular studies are required to ascertainsuch genetic changes. Yet, curiously enough, when comparativemolecular genetic studies are conducted, the morphology of manytumors and their DNA profile correspond closely to each other,and correlate with molecular genetic data and tumor behavior [11].Such studies are still uncommon but ultimately may lead to persuasiveresults of clinical value.

Unfortunately, many of the molecular genetic studies are conductedin a clinical vacuum without the close cooperation of cliniciansand pathologists. Consequently, data on the long-term clinicalsignificance of genetic changes often are lacking. Also, the resultsof some of these studies that receive a great deal of publicitymay prove to be only partially true or totally useless with thepassage of time.

Closer Cooperation Among Investigators Needed

It appears to this writer that a better cooperation among thevarious interested investigators may be helpful in addressingsome of the problems related to DNA ploidy and cell cycle analysisand its clinical significance. DNA measurements are easily accessible,relatively inexpensive, and useful in some clinical situations.They should receive wider attention among clinicians. To ascertainthe true biologic significance of these DNA changes may requiremany years of coordinated prospective studies by pathologists,molecular biologists, and interested clinical parties.

References:

1. Tribukait B: Flow cytometry in assessing the clinical aggressivenessof genitourinary neoplasms. World J Urol 5:108-122, 1987.

2. Tribukait B: DNA flow cytometry in carcinoma of the prostatefor diagnosis, prognosis and study of tumor biology. Acta Oncol30:187-192, 1991.

3. Fallenious AG, Auer GU, Carstensen JM: Prognostic significanceof DNA measurements in 409 consecutive breast cancer patients.Cancer 62:331-341, 1988.

4. Adolfsson J, Carstensen J, Lowhagen T: Deferred treatment inclinically localized prostatic carcinoma. Br J Urol 69:183-187,1992.

5. Forsslund G, Esposti PL: Prognostic significance of nuclearDNA levels in prostatic carcinoma. Scand J Urol Nephrol 55:53-58,1992.

6. Zetterberg A: Stability of diploid genome in carcinoma of theprostate with long follow up, in Andersson L (ed): Diagnosis andprognostic parameters in localized prostate cancer. Scand J Uroland Nephrol (supppl) 162, 1964.

7. Stephenson RA, James BC, Gay H, et al: Flow cytometry of prostatecancer: Relationship of DNA content to survival. Cancer Res 47:2504-2507,1987.

8. Winkler HZ, Rainwater LM, Myers RP, et al: Stage D1 prostaticadenocarcinoma: Significance of nuclear DNA ploidy pattern studiedby flow cytometry. Mayo Clin Proc 63:103-112, 1988.

9. Koss LG, Suhrland MJ: Atypical hyperplasia and other abnormalitiesof prostatic epithelium (editorial). Hum Path 24:817-818, 1993.

10. Agarwal V, Greenebaum E, Wersto R, et al: DNA ploidy in spindlecell soft tissue tumors and its relationship to histology andclinical outcome. Arch Pathol Lab Med 115:558-562, 1991.

11. Czerniak B, Cohen GL, Etkin P, et al: Concurrent mutationsof coding and regulatory sequences of the Ha-ras gene in urinarybladder carcinoma. Hum Pathol 23:1199-1204, 1992.