**Cervix Cancer Local Relapse rates based on surgical margins and use of post operative radiation**
Margins	No XRT	PostOp Radiation
negative	10%	0%
close (>0 - < 1cm)	17%	0%
positive (1cm +)	50%	25%

Cervix Cancer Local Relapse rates based on surgical margins and use of post operative radiation

Margins

No XRT

PostOp Radiation

negative

10%

close (>0 - < 1cm)

17%

positive (1cm +)

50%

25%

Influence of margin status and radiation on recurrence after radical hysterectomy in Stage IB cervical cancer Viswanathan IJROBP 2006;65:1501
Purpose: To examine the relationship between margin status and local recurrence (LR) or any recurrence after radical hysterectomy (RH) in women treated with or without radiotherapy (RT) for Stage IB cervical carcinoma. Methods and Materials: This study included 284 patients after RH with assessable margins between 1980 and 2000. Each margin was scored as negative (≥1 cm), close (>0 and <1 cm), or positive. The outcomes measured were any recurrence, LR, and relapse-free survival. Results: The crude rate for any recurrence was 11%, 20%, and 38% for patients with negative, close, and positive margins, respectively. The crude rate for LR was 10%, 11%, and 38%, respectively. Postoperative RT decreased the rate of LR from 10% to 0% for negative, 17% to 0% for close, and 50% to 25% for positive margins. The significant predictors of decreased relapse-free survival on univariate analysis were the depth of tumor invasion (hazard ratio [HR] 2.14/cm increase, p = 0.007), positive margins (HR 3.92, p = 0.02), tumor size (HR 1.3/cm increase, p = 0.02), lymphovascular invasion (HR 2.19, p = 0.03), and margin status (HR 0.002/increasing millimeter from cancer for those with close margins, p = 0.03). Long-term side effects occurred in 8% after RH and 19% after RH and RT. Conclusion: The use of postoperative RT may decrease the risk of LR in patients with close paracervical margins. Patients with other adverse prognostic factors and close margins may also benefit from the use of postoperative RT. However, RT after RH may increase the risk of long-term side effects.
This study analyzed the risk of recurrence in Stage IB cervical cancer patients after RH with histologically close paracervical margins (>0 but <1 cm between tumor cells and the specimen edge). Patients with close margins had a greater absolute risk of any recurrence (20%) than did patients with negative margins (11%). However, the risk of any LR for close margin status was more similar to the risk of those with negative margins (11% vs. 10%). The rate of LR was lower in all patients who received RT, regardless of margin status. However, this decrease was not significant given the small number of events in each subgroup. Most patients in this trial had Stage IB1 cervical cancer, with local and distant relapse rates similar to those previously published Several retrospective studies have reported on the results on patients with close or positive margins treated with postoperative RT . Many of those studies grouped patients with positive and close margins together. Others did not report the rates specifically for close margins and did not report the effect of an increasing distance from the tumor cells to normal tissue. We found that each millimeter increase in the distance from the tumor to the margin in patients with close margins significantly increased relapse-free survival on univariate analysis. This implies that it is important not only to avoid dissecting directly through tumor, but also to remove as much normal tissue as possible adjacent to the tumor at hysterectomy. Postoperative RT is routinely recommended for patients with high-risk features. However, the definition of “high-risk features” remains a topic of some debate. Our univariate results have shown that the significant risk factors for recurrence were the depth of invasion, positive margins, LVI, and tumor size. Similarly, a Gynecologic Oncology Group prospective trial identified the risk factors for recurrence after RH as tumor size, LVI, and depth of tumor invasion. Margins in that study were positive in 20 patients. The crude rate of freedom from recurrence was 84% for those with negative margins and 69% for those with positive margins. The difference was not statistically significant, likely because of the small numbers of patients. Subsequently, the Gynecologic Oncology Group conducted a randomized trial of hysterectomy with and without RT for early-stage cervical cancer patients with high-risk features. These features included at least two of the following risk factors: greater than one-third stromal invasion, capillary lymphatic space involvement, and large cervical tumor diameter. Overall, the risk of recurrence was reduced by 47% with adjuvant RT (p = 0.008). In the hysterectomy-only arm, 21% of the patients died during the study period vs. 13% in the postoperative RT arm Although the presence of positive LNs was not a significant predictor of recurrence in this study, the small number of patients in this subgroup limited interpretation of this as an independent factor. Other studies have clearly identified LN involvement as a significant negative prognostic factor. Koh noted a 34% local relapse rate in patients with positive LNs, and 25% had synchronous distant metastases. In the present study, LVI was identified in 60% of patients with LR. Univariate analysis showed that LVI was a significant predictor of relapse-free survival, but too few events occurred in the subgroup of patients with LVI to accurately calculate a multivariate risk estimate. Three of 25 reviewed articles identified LVI as an independent prognostic factor Postoperative RT must be given with caution, because it increases the overall risk of major complications, including small bowel obstruction. In the Gynecologic Oncology Group randomized trial, Grade 3 and 4 complications increased from 2% to 7% with RT after RH . In a study by Bandy, patients irradiated after hysterectomy developed more bladder contraction and instability than did those treated with surgery alone. In the present study, all acute and long-term side effects were increased in patients who received postoperative RT. This may, in part, have been related to the use of AP–PA fields in some patients. The impact of chemotherapy on LR rates with regard to margin status in cervical cancer has not been studied. In breast cancer, a study assessing the risk of LR with regard to margin status showed that chemotherapy reduced the risk for patients with focally positive margins to a level equivalent to that of patients with negative margins. Concurrent cisplatinum chemotherapy with RT in postoperative high-risk cervical cancer and those with positive margins is the standard of care. In our study, the small number of patients receiving chemotherapy limited analysis of this factor. We acknowledge some limitations of this study. This review was a retrospective analysis of the experience of a single institution with a small number of patients, particularly in the positive-margin subgroup. The small number of patients also limited conclusive statements about the role of RT. The results are therefore hypothesis generating. The selection criteria for local therapy were not consistently predefined as they would be in prospective trials. The small number of local events hindered our ability to report multivariate analysis results. A review of the side effects might have been biased because radiation oncologists may follow-up and record toxicities of treatment differently than do gynecologic oncologists. The group with close margins after RH represent a therapeutic dilemma for the radiation oncologist. The toxicity of postoperative RT must be weighed against the risk of local relapse and the ultimate goal of cure. Our results have indicated that postoperative RT may provide a maximal decrease in LR for patients with positive or close margins. Although RT decreased the absolute risk of recurrence in patients with negative margins, other adverse features should direct the use of postoperative RT in this subgroup to offset the potential risk of increased side effects.

Influence of margin status and radiation on recurrence after radical hysterectomy in Stage IB cervical cancer

Viswanathan IJROBP 2006;65:1501

Purpose: To examine the relationship between margin status and local recurrence (LR) or any recurrence after radical hysterectomy (RH) in women treated with or without radiotherapy (RT) for Stage IB cervical carcinoma.

Methods and Materials: This study included 284 patients after RH with assessable margins between 1980 and 2000. Each margin was scored as negative (≥1 cm), close (>0 and <1 cm), or positive. The outcomes measured were any recurrence, LR, and relapse-free survival.

Results: The crude rate for any recurrence was 11%, 20%, and 38% for patients with negative, close, and positive margins, respectively. The crude rate for LR was 10%, 11%, and 38%, respectively. Postoperative RT decreased the rate of LR from 10% to 0% for negative, 17% to 0% for close, and 50% to 25% for positive margins. The significant predictors of decreased relapse-free survival on univariate analysis were the depth of tumor invasion (hazard ratio [HR] 2.14/cm increase, p = 0.007), positive margins (HR 3.92, p = 0.02), tumor size (HR 1.3/cm increase, p = 0.02), lymphovascular invasion (HR 2.19, p = 0.03), and margin status (HR 0.002/increasing millimeter from cancer for those with close margins, p = 0.03). Long-term side effects occurred in 8% after RH and 19% after RH and RT.

Conclusion: The use of postoperative RT may decrease the risk of LR in patients with close paracervical margins. Patients with other adverse prognostic factors and close margins may also benefit from the use of postoperative RT. However, RT after RH may increase the risk of long-term side effects.

This study analyzed the risk of recurrence in Stage IB cervical cancer patients after RH with histologically close paracervical margins (>0 but <1 cm between tumor cells and the specimen edge). Patients with close margins had a greater absolute risk of any recurrence (20%) than did patients with negative margins (11%). However, the risk of any LR for close margin status was more similar to the risk of those with negative margins (11% vs. 10%). The rate of LR was lower in all patients who received RT, regardless of margin status. However, this decrease was not significant given the small number of events in each subgroup. Most patients in this trial had Stage IB1 cervical cancer, with local and distant relapse rates similar to those previously published

Several retrospective studies have reported on the results on patients with close or positive margins treated with postoperative RT . Many of those studies grouped patients with positive and close margins together. Others did not report the rates specifically for close margins and did not report the effect of an increasing distance from the tumor cells to normal tissue. We found that each millimeter increase in the distance from the tumor to the margin in patients with close margins significantly increased relapse-free survival on univariate analysis. This implies that it is important not only to avoid dissecting directly through tumor, but also to remove as much normal tissue as possible adjacent to the tumor at hysterectomy.

Postoperative RT is routinely recommended for patients with high-risk features. However, the definition of “high-risk features” remains a topic of some debate. Our univariate results have shown that the significant risk factors for recurrence were the depth of invasion, positive margins, LVI, and tumor size. Similarly, a Gynecologic Oncology Group prospective trial identified the risk factors for recurrence after RH as tumor size, LVI, and depth of tumor invasion. Margins in that study were positive in 20 patients. The crude rate of freedom from recurrence was 84% for those with negative margins and 69% for those with positive margins. The difference was not statistically significant, likely because of the small numbers of patients. Subsequently, the Gynecologic Oncology Group conducted a randomized trial of hysterectomy with and without RT for early-stage cervical cancer patients with high-risk features. These features included at least two of the following risk factors: greater than one-third stromal invasion, capillary lymphatic space involvement, and large cervical tumor diameter. Overall, the risk of recurrence was reduced by 47% with adjuvant RT (p = 0.008). In the hysterectomy-only arm, 21% of the patients died during the study period vs. 13% in the postoperative RT arm

Although the presence of positive LNs was not a significant predictor of recurrence in this study, the small number of patients in this subgroup limited interpretation of this as an independent factor. Other studies have clearly identified LN involvement as a significant negative prognostic factor. Koh noted a 34% local relapse rate in patients with positive LNs, and 25% had synchronous distant metastases.

In the present study, LVI was identified in 60% of patients with LR. Univariate analysis showed that LVI was a significant predictor of relapse-free survival, but too few events occurred in the subgroup of patients with LVI to accurately calculate a multivariate risk estimate. Three of 25 reviewed articles identified LVI as an independent prognostic factor

Postoperative RT must be given with caution, because it increases the overall risk of major complications, including small bowel obstruction. In the Gynecologic Oncology Group randomized trial, Grade 3 and 4 complications increased from 2% to 7% with RT after RH . In a study by Bandy, patients irradiated after hysterectomy developed more bladder contraction and instability than did those treated with surgery alone. In the present study, all acute and long-term side effects were increased in patients who received postoperative RT. This may, in part, have been related to the use of AP–PA fields in some patients.

The impact of chemotherapy on LR rates with regard to margin status in cervical cancer has not been studied. In breast cancer, a study assessing the risk of LR with regard to margin status showed that chemotherapy reduced the risk for patients with focally positive margins to a level equivalent to that of patients with negative margins. Concurrent cisplatinum chemotherapy with RT in postoperative high-risk cervical cancer and those with positive margins is the standard of care. In our study, the small number of patients receiving chemotherapy limited analysis of this factor.

We acknowledge some limitations of this study. This review was a retrospective analysis of the experience of a single institution with a small number of patients, particularly in the positive-margin subgroup. The small number of patients also limited conclusive statements about the role of RT. The results are therefore hypothesis generating. The selection criteria for local therapy were not consistently predefined as they would be in prospective trials. The small number of local events hindered our ability to report multivariate analysis results. A review of the side effects might have been biased because radiation oncologists may follow-up and record toxicities of treatment differently than do gynecologic oncologists.

The group with close margins after RH represent a therapeutic dilemma for the radiation oncologist. The toxicity of postoperative RT must be weighed against the risk of local relapse and the ultimate goal of cure. Our results have indicated that postoperative RT may provide a maximal decrease in LR for patients with positive or close margins. Although RT decreased the absolute risk of recurrence in patients with negative margins, other adverse features should direct the use of postoperative RT in this subgroup to offset the potential risk of increased side effects.