Making a ggsurvplot with a coxph

Question

I've coded the following coxphobject

cox <- coxph(Surv(INTERVAL, PX_STAT_CODE) ~ TX_NUM, data = outcomes_1and2above, robust = TRUE, weights = weights, cluster=subclass)

from a propensity score matched dataset and would like to create a KM plot that delineates by TX_NUM. I'm using the following code

fit <- survfit(cox, data=outcomes_1and2above)
ggsurvplot(fit, data=outcomes_1and2above)

which returns

I was able to do this previously with the unmatched dataset using the following code

outcomes_1and2above_Surv <- Surv(time = outcomes_1and2above$INTERVAL, event = outcomes_1and2above$PX_STAT_CODE)
km_fit_txnum_12a <- survfit(outcomes_1and2above_Surv ~ outcomes_1and2above$TX_NUM, data = outcomes_1and2above)
ggsurvplot(fit = outcomes_1and2above_Surv, data = outcomes_1and2above)

but am unable to replicate this with the matched data since the Surv function doesn't have a weights argument.