A Randomized Trial of Pelvic Radiation Therapy Versus No Further Therapy in Selected Patients with Stage IB Carcinoma of the Cervix after Radical Hysterectomy and Pelvic Lymphadenectomy: A Gynecologic Oncology Group Study
Sedlis A, ... Zaino RJ.
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 1, 2001
Reviewers: John Han-Chih Chang, MD
Background / Discussion / Critique
Early stage cervical cancer has a very good prognosis treated with either surgery (radical hysterectomy - RH) or radiation therapy (RT). The standard of care has been radical hysterectomy, despite evidence of similar outcome with RT. However, this is beyond the scope of this article. Stage IB lesions are defined as lesions confined to the cervix, but are grossly visible or palpable on exam. Stage IA is considered microscopic disease. Prognostic factors have historically been lymphovascular invasion (or CLS invasion in the article), anemia, young age, grade and bulk of tumor, and depth of stromal invasion. The authors cite a previous Gynecologic Oncology Group (GOG) study, which determined that the surgically treated patients had an increased chance of recurrence at 3 years from 2 to 31% if the patients had certain poor prognostic factors. One fourth of stage IB patients with negative lymph nodes at the time of surgery have those poor prognostic factors. With such a high recurrence rate and subsequent poor outcome, the GOG set out to establish a study comparing no further treatment after RH versus postoperative RT in poor prognostic patients as indicated above in the methods section. The median follow-up was 5 years.
Some minor noteworthy criticisms are the following. There were some imbalances in the distribution of the patients based on the prognostic factor make-up of each group. As demonstrated in Table 2 and 3, the NFT arm had more patients with CLS involvement. However, the RT arm had more patients with tumors greater than or equal to 4 cm. Supposedly, the authors performed an analysis that demonstrated that these skewed distributions did not bias the data. This would have been a cleaner trial had they stratified for the high-risk prognostic factors. The subjective "objective" determination of tumor size by the surgeon lends itself to interpreter bias. Because this is a multi-institutional trial, many different surgeons will apply their own judgment as to the clinical size of the tumor. A more accurate method may be to allow the pathologist to measure lesion size. Approximately 17% of the patients on the RT arm either did not receive any RT, got less than 85% of the intended (50.4 Gy) or had a greater than 20% protraction of their RT treatment time. Perhaps a larger difference in the improvement in survival and outcome may have occurred, if more of the patients had received the treatment intended for them. However, I do realize that this is always the major issue in randomized trials comparing one type of treatment to no treatment at all. Finally, the authors detailed the fact that of those who were considered alive, 8 and 3 were lost to follow-up within two years of enrollment in the RT and NFT arms. They were censored at the time of last follow-up, but they probably should be scored as failures to correctly document their outcome or have a separate analysis. The authors state that they did perform an analysis and that the difference in survival and disease control improvement in the RT arm remained significant, though no documentation was given.
Adjuvant pelvic radiotherapy following radical surgery reduces the number of recurrences in women with Stage IB cervical cancer at the cost of 6% grade 3/4 adverse events versus 2.1% in the NFT group.
Overall, the conclusion that the authors come to is that postoperative RT is indicated in poor prognostic RH patients with node negative IB cervical cancer. The cost is increased grade 3 - 4 morbidity on the order of 6% in getting RT versus 2% when you do not. This is the only randomized trial having demonstrated the survival benefit of postoperative RT in early stage cervical cancer. Despite its minor pitfalls, this GOG trial provides a very compelling argument to offer postoperative RT in these selected early stage cervical cancer patients with these poor prognostic indices.